← Back to Archives
VOLUME 15, ISSUE 4, APRIL 2026
Publication in progress..
DEVELOPMENT OF A CHILD ABUSE MONITORING AND REPORTING SYSTEM USING A NATURAL LANGUAGE PROCESSING MODEL
Chibuike Eusebius Nnaemeka, Obikwelu Raphael Okonkwo, Njideka Nkemdilim Mbeledogu
Real-Time Explainable Malware Detection with Automated Response
Mrs Ayesha Azeeza, Dr Nasreen Taj M
SOUNDFOREST: MONITORING FOREST BIODIVERSITY USING AI-POWERED SOUND CLASSIFICATION
MATHIR VISHNU S, REVATHI A
Predicting Customer Churn Using Advanced Machine Learning Ensemble Methods with Sentiment Analysis Integration
Sakthi Dharan S, Dr. A. Revathi
AI INTERVIEW PREPARATION SYSTEM
Harish Mythrayan T, Ashwin Shano S A, Baranidharan R
A Smart Blockchain-Based Platform for Fake Product Detection and Secure Product Verification
Khan Fardeen, Adnan Ansari, Khan Abusayma, Mansi Shimpi
SAFE DRIVE IOT: Intelligent System to Prevent Drunk Driving
Prof. Liril George, Mihir Kishor Kumbhar, Ravitej Hiremath, Soham Arun Chandorkar
Smart Waste Segregation and Monitoring System Using IoT
Prof. Rita Pawade, Vipul S. Manbhe, Shreya S. Wasnik, Kashish R. Shende, Tushar V. Gore, Govinda K Belekar, Yuvraj L. Wadaskar
MediCard+ An AI-Driven Smart Healthcare System for Centralized Medical Records and Decision Support
Prof. Manoj Babar, Sonali Patil, Purva Powar Patil, Yash Patil, Aneesh Patil, Pranita Yadav
Real-Time Human Emotion Recognition and Analysis using DeepFace and OpenCV
Shaikh Asif, Ayan Dawat, Sayyed Anas, Shaikh Mufeez, Shah Mohd Sharique
Brain Tumor Detection
Talha Tidhare, Chaudhary Ashfaque, Bagwan Affan, Shaikh Maaz, Imran Shahid Abdul Rasheed
The Impact of Artificial Intelligence and Digitalization on the Workforce: A Skill-Biased Technological Change and Human Capital Perspective
Harsh Bhatt, Amit Kumar Sahu, Mrs Harshita Gaikwad
Detection and Prevention of Lung Diseases Using Convolutional Neural Networks: A Comprehensive Review
Pradnya Meshram, Tanaya Kattekola, Yashashree Sankade, Sakshi Kature, Hitesh Chatur
Evaluating Customer Loyalty Program: Starbucks India
Somya Prasad, Vora Smit
AI-ENABLED NETWORK MONITORING SOLUTION
DR.D. VIMAL KUMAR, DR.A. REVATHI, K. ABIRAMI
PREDICTIVE ANALYTICS FOR WORKFORCE REDUCTION IN MULTINATIONAL CORPORATIONS: A LOGISTIC REGRESSION APPROACH
Dr. Vimal Kumar D, Mr. Arockiya John Prasanna X, Ms Hemalatha A
LegalBot: An AI-Based Legal Chatbot System for Accessible Legal Guidance
Bhoomi Singh, Rishabh Pandey, Bhoomi Gupta, Prarthita Lade, Mr. Kashif Shaikh
UPI Fraud Detection Using Hybrid Machine Learning Models with Explainable Risk Scoring and Real-Time Monitoring
Tarra Sekhar, G. Vijaya Kumar
Deep-SiamChange: A Multi-Scale Attention-Based Siamese Network for Robust Structural Change Detection in Urban Environments
D VIMAL KUMAR, A REVATHI, B YOGESHWARI
EnergyLint: A Real-Time Static Code Analysis Engine for Energy-Efficient and Sustainable Software Development
Satyam Pandey, Nilesh Vishwakarma, Suraj Patel, Arya Mayekar, Prof. Vaishali Rane
Sentiment Analysis Using RoBERTa-Based Hybrid Model
M. Srivalli, A. Sri Nagesh, Jaideep Gera, Sk. Moinuddin, P. Gnana Suji
An Efficient Image Classification System Using OpenCV and Deep Learning Models on Google Colab
Vishal Kumar, Kuldeep Chauhan, Varun Bansal, Sumika Jain
AI-Powered Multimodal Emotion Recognition: A Zero-Shot Framework Using Large Multimodal Models for Real-Time Affective Computing
Vanitha A, Mohammed Roshan Akther K M
Cloud-Based IoT System for Real-Time Monitoring and Filtration of Industrial Wastewater for Multi-Purpose Reuse
SK. Kalesha Vali, M. Nivas, Ch. R. V. Durga Sai, N. Sanjeeva Reddy, G. Vijay Kumar
Automated Real-Time Bottle Defect Detection Using YOLOv8, BoT-SORT Tracking, and Audio Alerts
SHENBAGA GANESHAN S, Dr. C. KARPAGAVALLI, Dr E. MARIAPPAN, Dr M. KALIAPPAN
An Intelligent AI-Based Vehicle Breakdown Assistance System with Network-Aware Mobile Deployment
KARTHIKEYAN M, Dr C. KARPAGAVALLI, Dr E. MARIAPPAN, Dr M. KALIAPPAN
Deep Learning Based Real-Time Sign Language Recognition
Prof. Minal Patil, Rhushabh Gaikwad, Rushikesh Ghogare, Shekhar Khandale, Roshan Avhad
Solar-IQ: An IoT Based Smart Solar Monitoring and Intelligence System
Rajshri Pote, Dhananjay Sharma, Vivek Sotra, Asmit Hade, Kapil Prajapati, Yash Chavhan
Railway Track Health Monitoring System Using IoT and ESP32: A Simulation-Based Approach on Wokwi Platform
Vladimir Illich Arunan V V, Ms. R. T. Charulatha
COMPREHENSIVE REVIEW: CANCER TYPE DETECTION USING ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING
Prateek Sikarwar, Saurabh Singh, Aman Singh, Rohit Sharma
A DATA-DRIVEN MACHINE LEARNING FRAMEWORK FOR EMPLOYEE PRODUCTIVITY CLASSIFICATION IN WORK-FROM-HOME SETTINGS
Dr. Angelpreethi A, Gayathiri S, P Anitha
Smart Gas Monitoring And Prediction System Using IOT And ML
M.Kumar, K. Mariya Priyadarsini
Digital Clinic Appointment and Scheduling System
Mrs. M. Khamar, S. Rajasri, S. Gayathri, V. N. V. Kamakshi, Y. Rohitha
Smart Expense Tracking and Tax Estimation Web Application
Mrs. K. Tejaswi, V. Yoshitha, T. Mounika, Y. Himeswari, V. Kavya
AI-Enhanced Student-Centric Learning Resources Catalogue and Management Portal with Integrated Chatbot Assistance
Mrs. J. Mounika, P. Nithin, S. Lokesh, P. Samuel Kowshik T. Siddharadha
Role-Based User Authentication, Authorization and Secure Content Delivery System
Mrs. V. Krishna Vijaya, V. Balaram, P. Srikanth, Y. Bharath, Sk. Roshan Zameer
Smart Action Oriented Student Feedback System
Mrs. B. Kalyani, V. Vardhan Babu, S. Syam Noel, P. Manideep Sai, R. Gowri Nikhil
User-Centric Bug Tracking Issue Logging and Resolution Workflow Management System
Mrs. K. Deepthi, Sk. Chinna Nagul Meera, P. Naga Babu, T. Gangadhara Rao, P. Venkata Hemanth Kumar
QR -Based Event Management System with an Attendance Tracking System
Mrs. K. Tejaswi, T. Srikanth, P. Srinivas, SK. Md. Yaseen, and S. Arun Kumar
Task-Oriented Productivity And To-Do Workflow Optimization Dashboard
Mrs. V. Krishna Vijaya, N. Mahesh, U. Abhinash, N. Bhairavanadh, N. Thrinadh Chowdary
Distributed Car Rental Reservation, Fleet Allocation and Customer Billing Management
Mrs. P. Saritha, N. Srinivasu, Sk. Mahabub Subhani, P. Sravan Kumar Reddy, V. Rambabu
Online Bookstore Product Catalog with AI-Powered Personalized Recommendations and Chat Support
Mrs. P. Aksha Dhanalakshmi, B. Lakshman, Sk. Sadath Ali, T. Sundar Dharahas, T. Karthik
Online Examination Question Bank, Quiz Attempt and Result Evaluation Portal
Dr. M. Purnachandra Rao, V. Srinivasa Rao, Thatha Hemanth, Surabi Sanjay, T. Chandu Vardhan
AI-Based Glove for Hearing Impaired
Poorna Chandra Tejaswi, Dr Kavitha A S, Nikhith Gowda R, Pavan S, Rakshitha S P
Artificial Intelligence in Healthcare: A Comprehensive Survey on Disease Prediction
Buddareddy Sumeeth, Akula Sai Venkat, E Madhumitha, Aditya Santosh Naik, Dr. Muhibur Rahman T.R
HYBRID DISENTANGLED GRAPH CONTRASTIVE LEARNING FOR INTENT-AWARE RECOMMENDATION
VANITHA A, REVATHI A, DARSHINI P.G
SMART ATTENDANCE BASED ON FACE RECOGNITION USING RASPBERRY PI AND GSM MODULE FOR HOME INTIMATION
N. Syam, T. Ravindra Sai, G. Chinna Jagapathi Swamy, K. M. Priyadarshini
WIRELESS POWER TRANSFER FOR ELECTRIC VEHICLES (EV) CHARGING IN MATLAB
S. Muralikrishna, Mr.M.Rama Krishna
Design and Development of Air Quality Monitoring and Alert Systtem using IoT
Dr. Ome Nerella, Tellakula Manideep
An EfficientNet-B4 Based Medical Deepfake Detection in Healthcare Image Analysis
Mr. A. Azeem, K. Gowthami, B. Indhu, K. Pavani
A Sensory Glove With a Limited Number of Sensors for Recognition of the Finger Alphabet of Polish Sign Language and with voice
T. Yatesh, P. Victor paul, Y. Veeranjaneyulu, Mrs. T. Manogna
EXPLAINABLE DEEP LEARNING-BASED INTRUSION DETECTION SYSTEM FOR WIRELESS SENSOR NETWORKS WITH REAL-TIME EDGE DEPLOYMENT
Chilka Sadhana, Mr K Appala Raju
A COMPARATIVE STUDY OF GENRE-BASED SENTIMENT ANALYSIS
Dr. Angelpreethi A, Hepshiba Sherly P A, P Anitha
Hybrid Adaptive Agentic Architecture for Autonomous Content Curation and Delivery: A Serverless Implementation for Real-Time Tech News Syndication
Yeshwathram D, Thaaiyumanavan G R, Vinod D
Advertising Across Eras Traditional vs Modern Media and Their Influence on Brand Equity
Manasvi Dubey, Gohel Roshani, Dr. Hiren Harsora
Voice Interview Analyzer
Mirza Daniyal Baig, Alfaiz Samani, Shehzan Shaikh, Anas Nakade, Alfiya Mulla, Zeeshan Khan
Intrusion Detection System for Smart Agriculture Using Deep Learning
Mr. M. Rama Krishna, M Tech, P. Keerthi Chandana, B. Varshita Lakshmi
Safe Park Sentry System using YOLOv8 Approach
Safe Park Sentry System using YOLOv, Approach, Abdul Sami Mansuri, Khan Ali Hamza, Yaheya Labbay, Koli Aradhya Umesh, Ali Naushadali Tangsal, Mr. Mohammed Sharique Maqsood Ahmed Shah
Development of Fluoride Detection and Alert System for Drinking Water
Khadar Basha Shaik, V. Akhilesh, S. Harshith, V. Rajeev Reddy, Dr. Ome Nerella
Early Disease Detection of Plants Using Deep Learning
THANGAMAHESWARAN V, Mr. P. VETRIVEL, Dr E. MARIAPPAN, Dr M. KALIAPPAN
Smart Highway
Prof. Suchitra Patil, Akhil Dandoti, Piyush More, Soham Pathare, Shubham Nagotkar
IoT-Based Industrial Machine Monitoring, Energy Analysis, and Safety Automation System
V. Yasaswini, G. Vijay Kumar
FitBuddy: A Personalized Digital Fitness Trainer for Health Monitoring and Lifestyle Management
Sumit Rothe, Sanskruti Pednekar, Samiksha Ovhal, Saloni More, Sandeep Shinde, Kanda Kumaran Thevar
Smart Agriculture System for Grape Leaf Disease Detection Using AI, Image Processing and Sensors
Prof. S. M. Bankar, Prof.Sweety Narula, Akshara R. Jadhav, Shravani G. Zurange, Bhagyashri V. Kalamkar
Energy Consumption Prediction in Smart Homes using IoT and Machine Learning
Aditi Srivastava, Aditi, Anshika Tyagi, Chhavi Garg, Diya Jain, Usha Kumari, Satish Kumar Soni, Uruj Jaleel
SolarMap AI: An AI-Based System for Personalized Solar Planning
Rajeshree Chaudhari, Dnyaneshwari Sonawane, Sania Shinde, Prof. Satish Kuchiwale
Civic Issue Reporting and Tracking
Sanskruti Maskar, Purva Patil, Rajshri Chavare, Vaishnavi Bharmal, Mr. Akshay Jeevankumar Shinde
IoT and Machine Learning Based Smart Solar Energy Monitoring and Automatic Dual-Axis Solar Tracking System
G.Ravi, B.Lohitha
IoT and Machine Learning Based Smart Health Monitoring System
Sidagam Geethika, P. Bose Babu
SOLAR ROOF TOP WIRELESS CHARGING STATION FOR ELECTRICAL VEHICLES
T.Anjaneyulu, G.Ravi
BabyEase Cradle: Automated Infant Care System
Prof.Sweta Waghmare, Atharv Shirkar, Saeed Thakur
Explainable AI Driven Multimodal Framework for Robust Spinal Muscular Atrophy Detection
Smit Mahesh Wani, Namdeo Baban Badhe, Neeta Patil
Trustify: An Intelligent Multimodal Framework for Fake Review Detection with Explainable AI
Shraddha Shirish Shah, Dr.Anil Vasoya, Pranjali Kasture
COMPACT SELF-DIPLEXING MIMO ANTENNA WITH IMPROVED ISOLATION CHARACTERISTICS
Mr. N. Bujji babu, K. Karthik, L. Sai, K. Sridhar, K. Kumar Raju
IoT-BASED COAL MINE WORKER SAFETY AND ENVIRONMENT MONITORING SYSTEM WITH CLOUD ANALYTICS
K. Chaitanya Varma, Mr. M. Rama Krishna
AI ASSISTED AUTOMATIC TAMARIND PULP EXTRACTOR FOR DOMESTIC SCALE
Mrs S.Nila, S.Divya, R.Deenadalayan, R.Mothish
Diabetic Retinopathy Classification Using Vision Transformer: A Strategy for Small Dataset Challenges
Mr K. Appala Raju, M. Pravallika, M. Bhumika, M.V.K.S. Harika
EcoReport Civic Complaint Management System
Lect. Satyajit Deshmukh, Vishwajeet Ukey, Avinash Chauhan, Ritick Raj Pandey, Md Islam, Abdul Slam, Dhruv Patel
BRAIN-COMPUTER INTERFACE SYSTEMS USING ARTIFICIAL INTELLIGENCE: A REVIEW OF EEG-BASED APPROACHES
Vidya K T, Rajeshwari N
VISIBILITY ENHANCEMENT OF LESION REGIONS IN CHEST X-RAY IMAGES WITH IMAGE FIDELITY PRESERVATION
Mr N. Bujii Babu, Koilapu Preetham
Pipeline Maestro: Intelligent Orchestration of Build, Test and Deploy in CI/CD Systems
Prof. Diksha Bansod, Riya Mundrika Patel, Trupti Narhari Khairkar
SMART INTRUDER DETECTION
Shree Rakshana R, Akrithi RJ, Rashmi R, Ms. Charulatha. R.T
Personalized Skin Disease Consultant and Care Recommendation Using Lifestyle-Based Analysis
Mrs. S N Khandare, Pratik Birpan, Anup Sawai, Mayuri Pathade, Akanksha Gurav
AI-Based Research Paper Analyzer And Generator Using NLP And Machine Learning
Prof. Diksha Bansod, Shubham Gokul Jadhao, Himanshu Gulab Dhande
AUTOMATED RAILWAY SIGNALING SYSTEM WITH LOOPLINE FEEDBACK
Mr N. Bujii Babu, Janapala Balu
Lung Cancer Detection Using Convolutional Neural Networks
P. Sampath Kumar, Dr. S. Mallikharjuna Rao*
An Optimized IoT-Based Coal Mine Safety Monitoring System with Edge-Driven Real-Time Hazard Detection
Bhavana L, Adithya Pratheep, Sarah Ninan, Vidya Sree, Shruthilayaa. M, Ms Charulatha R.T
Security Issues in Electric Vehicle Charging Infrastructure
Ms. Disha S. Gopatwar, Mr. Kshitij P. Tajne, Mr. Piyush S. Borkhade, Prof. A. A. Gophane
Special Minds Companion Assistant Using AI
Parth Khot, Pranav Akiwate, Rohit Babar, Pradnya Katkar, Shravani Jadhav, Shruti Hogade, Prof. K. S. Kadam
AI-Enhanced SMS Spam Detection Using Hybrid NLP and Machine Learning Techniques
Manish Singh Gahlot, Divyanshu Sharma, Bharat Saini, Dhananjay Kumar, Gagan Sharma, Ajit Singh, Satish Kumar Soni, Uruj Jaleel
AUTOMATED VERTICAL GARDENING SYSTEM USING IoT-ENABLED SMART IRRIGATION FOR SUSTAINABLE URBAN GREEN SPACES
Dr. Bhagyashree Dharaskar, Chandan Lokhande, Mohit Tembhurne, Kunal Patle, Mohit Goupale, Gaurav Hiwase
SmartBin: An IoT-Based Intelligent Waste Classification and Automated Segregation System Using Machine Learning
Shri Nivetha R, Kavya S Prasad, Jeyaarthi V, Hasanthika Sri S, Deepashri P M, Ms Charulatha R T
Explainable Artificial Intelligence-Based Network Intrusion Detection System Using SHAP, LIME and Counterfactual Analysis
Manam Siva Sai, Dr. Chandra Sekhar Koppireddy
Artificial Intelligence Governance and Policy Analytics using NLP
Ms. Neha W. Bandabuche, Dr. V. H. Deshmukh, Dr. Y. A. Dhumale
IoT and Machine Learning Based Air Quality Monitoring and AQI Prediction System
S. Pujitha, M. Rama Krishna [M. Tech, PhD]
Automated Soybean Crop Health Evaluation from UAV Images Using Patch-Level CNNs
Saket Bobade, Sangram S. Dandge, Dr. Vaishali H. Deshmukh
Enhancing Shared Mobility Through AI-Driven Bike Rental Platforms
Prof. Diksha Bansod, Aaditi Katole, Janvi Aher, Sumit Ghoshal, Jay Ingole
Design and Development of Drinking Water Quality Alert System with Voice Announcement for Remote Areas
A. Haresh Kumar, B. Bhartha Kumar, G. Ganga Reddy, Dr. Ome Nerella
DeskMind: A Modular Desktop-Based AI Assistant for Voice and Text Command Processing
Lajari Atul Patil, Aditya Ganesh Poojari, Parth Kishor Rane, Siddhesh Deepak Wagh,Dr. Avinash Sonule
Hybrid Deep Learning and Transformer-Based Approach for Accurate Email Spam Classification
Dhanashri Shukla, Suyash Shrivastava
A SYSTEMATIC REVIEW: OF AI-BASED MENTAL HEALTH CHATBOTS: TECHNIQUES, CHALLENGES, AND FUTURE DIRECTIONS
Aman Kumar Sharma, Amit Kumar, Ashish Solanki, Ashish kumar, Aditya Sikarwar, Alok Singh Jadaun
MINEWATCH: LORA-ENABLED COAL MINE MONITORING WITH RASPBERRY PI PICO
U.Vamsi Krishna, K. Appala Raju
SOLARFLARE: AI AND IoT FUSION-BASED SOLAR PANEL PROTECTION SYSTEM
M.Rama Krishna, Sk Shafiulla
DESIGN AND DEVELOPMENT OF AN ERGONOMIC BULLOCK CART YOKE USING CAD MODELLING AND FINITE ELEMENT ANALYSIS
Atharva Bhedse, Hariom Jat, Anirudh Menon, Raj Patil
Intelligent Energy Optimization of Electric Vehicles Using Coordinated Regenerative Braking and Advanced Control Strategies
Tulsi D. Kuralkar, Aditya G. Kale, Disha S. Gopatwar, Prof. A. A. Gophane
Smart Waste Management System Using Deep Learning
Mr. M. Rama Krishna P. Rupas
HERB GUARD AI- AN AI - BASED HERB-DRUG INTERACTION PREDICTION USING NATURAL LANGUAGE PROCESS
Shashipriya Shridhar Hegde, K Sharath
AUTOMATION DETECTION OF STEGANOGRAPHICAL CONTENT USING MACHINE LEARNING
Laxman Bhandarwad, Yogiraj Deshmukh, Nitesh Jadhav, Dr Taware G.G
AI-Based Violation Alert System for Helmets
Chandan KL, Dr Kavitha A S, Jeevan Pranav D, Punith HM, Rahul R
Accessibility in Fitness Technology: Ensuring Inclusive Digital Health
Rachit Agarwal, Vishesh Vats, Kirtika Saini, Shikhar Chauhan, Ajit Singh, Dr. Uruj Jaleel, Dr. Satish Kumar Soni
SMART STOVE SAFETY AND MONITORING SYSTEM
Mrs S. Dharani, Aesteen Abraham, Charumathi. B, Remna R.S
Design And Development of Solar Powered Intelligent Weed Removal system for Sustainable Farming using ESP32
Mrs.Affrose, Padigelawar Hemanth, Vatrapu Sathya Sai Reddy, Routu Girish
Cloud Guardian: An AI-Assisted Edge-Based AWS Audit and Optimization System
Mr. E. Lingamurthy, Md Majid, P. Mallikarjun, M. Yashwanth
A Multi-National Framework for Real-Time Collaborative Cyber Defense: Evaluating FL Architectures and Aggregation Strategies in Heterogeneous NIDS
Abhiram T Sajeev, Adarsh S J, Alen J S, Alfin Jerome, Amila A L
PROSE: Prompt Refinement, Optimization and Semantic Evaluation
Chetan Kokate, Vedant Khadye, Shubham Borate, Chaitanya Kokate
RFID-Based Attendance System: Design, Implementation, And Performance Analysis Using ESP32 Microcontroller With RC522 RFID Module And I2C LCD Display
Ms.Charulatha. R.T, Thanish L. S., Sushaanth.M, Milan Prabhu.T, Varun. V.V, Gokul.J
CEREP: A Graph-Constrained Explainable Reasoning Engine for Multi-Omics Precision Oncology
Sushmita Kundu, Dev Mehta, Charulatha. R. T
ClashVerse: A Real-Time 1v1 Coding Battle System and Secure Code Execution
Nitin Gupta, Soham Dalvi, Prathmesh Ughade, Mayur Chaudhari, Shilpali Bansu
E-Commerce With Bidding System
Prof. Bina R. Rewatkar, Priyanshu P. Narayane, Shraddha M. Khodankar, Purva M. Mangrulkar, Anuj A. Kotangale
A Remote Sensing Nano-Satellite Time Control and Communication System
Hamsika M K, Inchara Jain B J, Janhavi M S, Ramya K
E-Sports Mobile Application Using Flutter and Firebase
Prof.Diksha Bansod, Aditya A. Langade, Amit S. Kevat, Anurag T. Gajhbiye, Ashvaghosh A. Gaurkhede, Pranav S. Machave
Design and Development of Intelligent system for Water cleaning Robot using IoT
S. Shiva Sai, Y. Shiva Reddy, Y. Akhil, Mrs K. Rashmi
ADAPTIVE INTELLIGENT LEARNING SYSTEM USING LARGE LANGUAGE MODELS AND LANGCHAIN
Ms.Vedanti U. Deshmukh, Dr. S. P. Akarte, Dr. G. R. Bamnote
Availability-Aware Multimodal Deep Learning for Breast Cancer Diagnosis with Missing Modalities
Indu P. K, Dr. G Beni, Dr. D Rene Dev
A Hybrid Signal-Processing and Neural Network Pipeline for Low-Latency Acoustic Event Detection at the Edge
Jatin Verma, Abhishek Jatla, Thanuja Reddy, Ms. Charulatha R.T
Multilingual AI-Based Voice-Controlled Robotic System Using Distributed Architecture
Mrs. V. Divya Vani, Dr. G. Anand Kumar, K. Dharan, Ch. Tharun
Enhancing Social Network Prediction in Graph Neural Networks Using Graph Theoretical Approaches: Social Network Analysis on The Facebook Ego Dataset
Pharsana Parveen M, Sr. Stanis Arul Mary A
IOT Based SOS Smart Helmet with Automatic Accident Detection and Real – Time Location Alert System
Akash Kumar, Sudharsan J, Jeron Sam Varghese, Sanjith TS, Hariharan N, Ms. Charulatha R.T
LUMEX: A Web-Based Image Enhancement Studio Using Python, Flask, and Pillow
Prataparao Charith, K.Varun, Yash Kumar, Dr. H. Mary Shyni
Gesture Talk
Tejas Sawant, Gaurav Singh, Shubham Tiwari, Aman Verma, Dipali Shende
Course Cloud: AI-Based Video Description Generator for Smart E- Learning
Shivam Giri, Shivani Kashyap, Shobhit Sharma, Udit Tyagi, Usha Kumari, Dr. Uruj Jaleel, Dr. Satish Soni
“Explainable Bone Tumor Diagnosis Using Deep CNNs and Language Models”
Prof. Amit Meshram, Vishal Suresh Nemade, Abhishek Sudam Pawar
Design and development of Sewage Drain Monitoring and alert system using IoT and LoRa Technology
P. Shanmukha Kumar, T. Srujan, Mohd. Faraz
Real-Time Face Emotion, Hand Gesture Recognition and Voice Conversion System for Deaf and Dumb
Nehan Naveen, Akshay Raj, Gopika C, Sibi D
An Intelligent Video Surveillance System for Multi-Threat Detection and Real-Time Alerting
Aniket Ligam, Avadhoot Katte, Atul Shelar, Niharika Dasari, Rakesh Suryawanshi
Fuzzy Inference Systems for Optimized Drug Dosing in Heart Failure Management
Varsha*, Prof. Diwari Lal
A Robust Offline-First Architecture for Integrated Examination and Institutional Data Management in Rural Educational Societies
Prof. Diksha Bansod, Shruti Bangare, Vinit Pawankar
AN OVERVIEW ON: AI STOCK FORECASTER: A STOCK PRICE FORECASTING APP
Prof. Priya Farkade, Uday Lanjewar, Ranit Garude, Mohan Khobarkhede, Rohit Bawanukey, Manthan Ukey
A STUDY OF PERSONAL FINANCIAL PLANNING AND SAVING BEHAVIOUR OF SALRIED EMPLOYEES 1Shraddha Kadam 2Nikita Khandade 3Janhavi Kasar 4Omkar Khutwad
Dr. Pradyuman Shastri
AN OVERVIEW ON: MEDICAL SYMPTOMS CHECKER APP
Prof. Sonal R Tiwari*, Sagar Kamble, Vaibhav Gawade
AN OVERVIEW ON: AI MEETING INTELLIGENCE A MEETING ANALYZER APP
Prof. Atul Akotkar, Yogesh Puri, Yash Gadling, Shatayu Meshram, Shivdas Lakhe, Piyush Chauhan
AI - POWERED HUMAN RESOURCE MANAGEMENT SYSTEM (HRMS)
Bina Rewatka, Rohit Umredkar, Vaishnavi Tandekar, Priyanka Choudhari, Shrishakti Sharma
Conflict-Free Replicated Data Types: An Exhaustive Analysis of Theoretical Foundations, Synchronization Protocols, and State-of-the-Art Architectures
Vivek R, Panchami M Hegde
Effect of Image Pre-Processing Techniques on Object Detection
Aparna M, Dr. H Mary Shyni
BIKE ON RENT: A WEB-BASED BIKE RENTAL MANAGEMENT SYSTEM
Atul Akotkar, Reena Kakde, Himanshu Nagpure, Nikita Dhote
Smart Traffic Management Web System Using AI and Real-Time Analytics
Varsha Santosh Ekhande
A Severity-Aware Hybrid ML Model for Real- Time Cyber Threat Detection and Alerting
Bhavani Kothapalli, Bodapati Preethi, Brahma K, K P S Kavya
AI-Based Smart Healthcare Assistance System
Prof. Diksha Bansod, Sneha K. Shrirame, Triveni M. Kirsan, Pranav S. Machave, Payal A., Uikey, Aditya A. Langade
Multimodal Deepfake Detection Using Deep Learning
Dhanushree B V, Panchami M Hegde
Virtual Reality–Integrated Telepresence Robot for Intelligent Remote Monitoring
Dr M V Sreenivas Rao, Jia M S, Preethu K
CyberWatch: A Real-Time Intelligent Web Security Monitoring and Attack Detection Platform
Drupad Dhamdhere, Shardul Pharande, Aashish Kondgire, Sakshi Phutane, Prof. Supriya Jagtap
POTHOLE & CRACK DETECTION SYSTEM USING ML & COMPUTER VISION
Prof. Amit Meshram, Komal Rewaskar, Pratiksha Tidke, Tannu Rangarkar, Akshada Sable, Tanushree Dhote
Development of a Piezoelectric Footstep Power Generation System with Arduino-Based Monitoring
Karthi Balaji M, Suraj K, Rudrakash Agarwal, Nithila K, Sandhya Dillybabu, Ms Charulatha R.T
IOT PROJECT: PEOPLE COUNTER USING IR SENSORS
SREEJITH S, DURAI RAJ R, MANISH KUMAR MANDAL, ARAVIND SRIRAM, DEEPAK G, MS . CHARULATHA R T
INDUSTRIAL FAULT MONITORING SYSTEM
Mr. M. RAVI, M. Tech(Ph. D), T. VENU MADHAV*
IoTCaloTrack: An IoT-Enabled Real-Time Calorie Tracking System
Akshat Kumar, Sauban Tarique, Preetam N, Yogesh P, Tharun VS, Ms. Charulatha R.T
Audio Video Dubbing System
Saakshi Nalawade, Shruti Namaye, Sahil Raut, Shreyash Sawant, Manoj M. Deshpande
A Deterministic Multi-Metric Framework for Automated Image Dataset Validation in Computer Vision
M.Balavignesh, Dr. C. Karpagavalli, Dr. M. Kaliappan
Comprehensive Analysis of Proactive Fraud Prevention and Contribution of Machine Language
Harshit Sharma, Himanshu Chauhan, Muskan Sharma, Mayank Sharma, Dr. Satish Kumar Soni, Dr. Uruj Jaleel
A STUDY ON MATERIAL DISPATCH AND DOCUMENTATION PROCESSES IN A MANUFACTURING COMPANY
Anish Nivas Shidruk, Dr. Pradyuman Shastri
POINT OF USE & POINT OF ENTRY IN WATER TREATMENT
Prof. Nikhil Gahukar, Prof. Vrushali Paraye, Prof. Sameena Ansari, Prof. Mohammad Rafiullah, Prof., Swati Jadhav, Honey A. Bakshi, Dipti V. Bawane, Vaibhav L. Dhurve.
AI-Driven Gesture Tracking: A Comprehensive Review of Techniques, Applications, and Future Directions
Huda Khan, Isha Kaushal, Gunjan Rani, Nikhil Kumar, Mr. K.S. Mishra
EDUCONNECT PORTAL: A WEB-BASED COMMUNICATION SYSTEM FOR PARENTS, TEACHERS AND ADMINISTRATORS
P.P Garate, S.K Gavli, V.A Ghag, V.S Gaikwad, V.A Bhamre
Evolutionary Deep Learning Approach for Driver Drowsiness Detection Using CNN and Genetic Algorithm
Bellamkonda Venkata Sudheer kumar, Dr. Shaik Javed Parvez
AROS: An Intelligent Arctic Route Optimization System Integrating Machine Learning with Environmental Risk Evaluation
Tejas Mali, Abhay Kulkarni, Varun Rasal, Ritik Gohil, Smita Chunamari
Study of Design and Mobility Approaches in Rocker–Bogie Suspension Systems
Sarthak Sharma, Utkarsh Nehwal, Yash Kaushik, Shrikant, Sushant Dhasmana, Dr. Uruj Jaleel, Dr. Satish Kumar Soni
TradeSphere: A Full Stack Web-based Stock Trading Platform Using MERN Stack
Aaditya Gupta, Akshay Dhiman, Aman Kumar, Harsh Sharma, Dr.Uruj Jaleel, Dr.Satish Kumar Soni
Land Use and Land Cover Classification Using Sentinel-2 Satellite Imagery
𝐏𝐫𝐨𝐟. 𝐀𝐦𝐢𝐭 𝐌𝐞𝐬𝐡𝐫𝐚𝐦, 𝐃𝐡𝐚𝐧𝐬𝐡𝐫𝐢 𝐃𝐮𝐤𝐚𝐫𝐞, 𝐒𝐚𝐧𝐠𝐡𝐚𝐫𝐚𝐭𝐧𝐚 𝐏𝐚𝐭𝐢𝐥, 𝐑𝐢𝐭𝐞𝐬𝐡 𝐋𝐨𝐧𝐚𝐫𝐞
Smart Traffic Management System Using Machine Learning for Adaptive Signal Control and Emergency Vehicle Prioritization
Aditya Nanote, Srujal Ogale, Vikrant Palvi, Lokesh Patil, Prof. Rakesh C. Suryawanshi
SIGN LANGUAGE ANALYSIS USING ARTIFICIAL INTELIGENCE
Aarti Dhage, Sanika Pawar, Vaishnavi Tilekar, Kanyakumari Mangrule, Prof. Dr. Sachin Bere, Prof. Mr. A.M Suryawanshi
Optimized Travel Recommendation & Itinerary Generation
Nidhi Dhake, Shravani Mestry, Akshay Diwate, Nishant Gudade, Dr. Siddharth Hariharan
SPARK-A Dating Application: A JAVA, Flutter and ANDRIOD-Based
Prof. Madhuri Parate, Ishant Fulzele, Yash Jugseniya, Gaurav Katole, Aditya Rewatkar, Rutuja Bhende
AN OVERVIEW ON: CREDIT RISK ANALYSIS
Prof. Amit Meshram, Bhagyashree Kohad, Saloni Chitalkar, Pranali Ganvir, Nikita Adhau, Himanshu Tadas
Healthcare Android Predictor
Ashwini Vatne, Pallavi Shrimangle, Dipali Lahane, Prof. D. S. Jaybhay
Optimized Automatic Timetable Generation Using Genetic Algorithm for Educational Institutions
Yash Gadekar, Ganesh Gaikwad, Diksha Gaikwad, Prof. Veena Bhamre
The Impact of AI on Financial Decision Making in The Automobile Industry
Dhanalakshmi G, Dr Lumina Julie R
A Study on Hr Policy Development and Data Management in Biotechnology Sector
Jayakumar R, Dr Tamilselvan P
Impact of Buy Now Pay Later Services on the Spending Behavior of Generation Z Consumers in Chennai
Karthikeyan S, Dr.Lumina Julie R
AI-Based Expense Tracker for Personal Finance Management
Kanishka, Prachi Pundir, Nandini Kumari, Mayank Kumar Roy, Uruj Jaleel, Satish Kumar Soni
DIGITALIZATION OF HR OPERATIONS IN THE SECURITY SERVICES SECTOR
Nithish K R, Dr. Geeta Kesavaraj
Impact of Corporate Social Responsibility on Community well-being among IT Industry.
Prabha C, Dr.Lumina Julie R
A STUDY ON LOGISTICS OPTIMIZATION ON MILK COLLECTION AND DISTRIBUTION
R. Dhanush, Kokila.K
AN OVERVIEW ON: CREDIT RISK ANALYSIS
Prof. Madhuri Parate, Meena Godghate, Hanisha Bulhe
MANAGING STRESS IN THE WORKPLACE THROUGH HR PRACTICES IN IT INDUSTRY.
Robina A, Dr. R. Lumina Julie
A Study on Impact of Social Media Marketing Strategies on Consumer Buying Behaviour
Ms S. Mithra, Dr. M. Rajapriya
A STUDY ON IMPACT OF WEBSITE USER EXPERIENCE CONVERSION RATES
Seethaladevi. S, Dr. Felisiya.M
Car Accident Detection with Alert System
Gayatri Mali, Shruti Desai, Vaibhavi Mhaske, Chaitanya Sonawane, Prof. Mrs. Dhage T.S, Mr.Bhapkar A.S
A STUDY ON SEO STRATEGY DEVELOPMENT AND IMPLEMENTATION FOR SMALL BUSINESSES: A CASE STUDY OF WEBOIN TECHNOLOGIES PVT LTD
Thangavelan K, Dr. Felisiya. M
A STUDY ON SOCIAL MEDIA MARKETING AND THE IMPACT OF ONLINE PROMOTION WITH REFERENCE TO WESTSIDE
M. Udheepa Reddy, Dr. Felisiya. M
Fruit Detection and Classification Using CNN with Deep Transfer Learning
Prerna, Sejal Rana, Dr. Satish Kumar Soni
SAHA -Women Safety Application and Device
Vedangi Shastrakar, Anamika Shirke, Sanika Shirsat, Parita Thakur, Dr. M.M. Deshpande
Secure Image & Audio Steganography Using AES-256 Encryption and Hybrid DWT-SVD Frequency Domain Embedding
M. Tejas Srinivasan, S. Roshan Pranao, Y. Sai Dheeraj, Ms. N. Saranya
Leveraging AI for the Next Era of Precision Oncology in Breast Cancer Detection
Amritpal Singh Yadav*, Virendra Kumar Sharma
AshaConnect: Bridging Healthcare to Every Doorstep
Prapti Borkar, Sakshi Bagul, Tanmay Ghodke, Bhavesh Dhangar
Immersi-Lab: Empowering Disabled Learners Through Virtual Biology Simulations
Faheem Ahmad, Dev Verma, Tanya Verma, Harsh Sharma, Dr Uruj Jaleel, Dr.Satish Kumar Soni
A SURVEY OF AI-DRIVEN INTRUSION DETECTION SYSTEMS FOR CLOUD AND EDGE COMPUTING: TECHNIQUES, CHALLENGES, AND FUTURE DIRECTIONS
Abdulrahman Mohammed Saba, Alfa Muhammad, Sayuti Musa Shafi’i
AN OVERVIEW ON: CAMPUS HIRE
Prof. Frah Nikhat, Trupti Karemore, Shruti Ruikar, Umair Qureshi
AI - DRIVEN INNOVATIONS IN MUTUAL FUND ANALYTICS
SANKAR C, Dr. S. ARUL KRISHNAN
A STUDY ON THE FACTORS INFLUENCING CANDIDATE DROPOUT IN RECRUITMENT PROCESS
Anandavathini K, Dr. Raja. S
Role of Social Media in Creating Brand Awareness among Rural Farmers
Aishwarya. S, Dr. S. Arul Krishnan
ScholarGrid: A Full-Stack Academic Collaboration Platform with Community-Driven Quality Layers
Ranvir Wadhawan, Swaroop Mandal, Aryan Yadav, and Meet Patil, Prof. K.L. More
“Next Generation Learning: Augmented Reality in Smart Education System”
Shivam, Priyanshu Ahlawat, Riya Sharma, Tanu Gupta, Rahul Vashishth, Miss. Taniya Jain, Dr. Uruj Jaleel, Dr. Satish Kumar Soni
AI-Based Spam Message Shield with WhatsApp Message Detection using NLP
Arti Kumari, Akansha Sharma, Ankita Katheria, Rooban Agrawal, Satish Kumar Soni, Uruj Jaleel
The Role of Nmap in Modern Network Security
Saurav Saini, Nikhil Singhal, Shekhar, Tanya Jain, Uruj Jaleel, Satish Kumar Soni
An AI-Automated Diagnosis System for Pneumonia Using Xception CNN
Aruna S, Deepika S, Harini M, Maheshwari B
IntelliInterview: An AI-Based Interview Training Platform Using Natural Language Processing
Aditya Kumar, Amisha Jain, Akshit Kumar, Ajit Singh, Dr. Uruj Jaleel, Dr. Satish Kumar Soni
Image Recognition Using Convolutional Neural Networks
Krish Kumar, Ujjwal Panwar, Mohit, Ajit Singh, Satish Kumar Soni, Uruj Jaleel
Web-Based Explainable Credit Card Fraud Detection Using SMOTE, Ensemble Feature Selection, and XGBoost
A. Ruba, Hema Lakshmi L, Vasumathy A
BorrowBridge: A Community-Based Shared Resource Platform for Sustainable Consumption
Aparna Chavan, Tanmay Hursale, Aayush Jadhav, Prof. Shilpa Tandale
Language Agnostic Conversational Intelligence System For Smart Campus
M Chandana¹, Parimala c², R M Harshitha ³, Saniya Hundekar ⁴, Dr.Muhibur Rahman T.R5
Three Phases Fault Detection System
Dr. Rahat Ullah Khan, Himanshu Verma
REAL TIME SUSPECT DETECTION AND TRACKING USING AI BASED SURVELIIENCE SYSTEM
Jeffy John Binu C, Iniyan R G, Baskar P N, Mohammad Adhil KV A
Home Blood Sample Collection System: A Smart Health Portal for Home Diagnostics
M Tharun Kumar, Rohith Edwin, ShivKumar Gowda, Sachin E, Dr. Muhibur Rahaman T.R
AI Driven Railway Track Crack Detection And Classification With YOLOV
Koppisetti Sriram, Gonuguntla Brunda, K. Akila
Sports Management System
Nitin Chaudhari, Mayur Gaikwad, Girish Kale, Amit Kanojiya, Prof. Madhuri Dalal
Crowd Abnormal Behaviour Detection Using YOLO V
Punitha N, Vasu Devan B, Sundhara Chozhan K, Tamil Selvan B, Thavasi M
ATS Resume NLP Analyzer: A Hybrid, Explainable, and Practical Framework for Resume-Job Matching
MD Auranzeb Khan, Srijan Mani Tripathi, Kunal, Durga Devi
An intelligent immigration and safety awareness platform
Aditya Raman, Vyom Pandey, Mohammad Rayyan Basha
A Context-Aware Security Layer for Preventing Account Takeover Attacks in Online Transactions
N.punitha, ME., Omeshwar K, Poovarasan K, Sampath S, Shahul Hameed N
A Machine Learning-Based Precision Agriculture System for Crop Recommendation and Yield Prediction
Alok Kumar Dwivedi, Harshit Tiwari, Yogendra Kumar, Kumar Bibhuti Bhushan Singh
Edu-Vision: An AI-Powered Multimodal Educational Assistant for Intelligent Content Understanding and Personalized Tutoring
Sura Reddy, George A, Abhishek M, Dr.Paavai Anand
IoT-Based Smart System for Arthritis Pain Monitoring and Relief
Abinaya S, Bavadharani R, Gajalakshmi P, Katherine R
DEEP LEARNING FRAMEWORK FOR SUPER ENHANCER PREDICTION USING CNN AND MULTI-HEAD ATTENTION WITH CROSS-SPECIES TRANSFER LEARNING
Parimala M, Naveen A, Veeradinesh R, Vikram M, Vinoth G
VIRTUAL TRY-ON FOR CLOTHES USING OPENCV
Bavip S, Dhanush S, Gokul Kumar S, Mokesh Raj V, Ms. MADHULIKA MK., M.E., Dept of Artificial Intelligence and Data Science
Multilingual Summarization of Youtube Video using NLP
Ms.KaviPriya M.Tech, Bharath S, Jayaganesh S., Kishor S, Mithun S
BUILDING MANAGEMENT SYSTEM
Vatsal Jain, Darshan Kandalgaonkar, Gautam Enjam, Deepesh Ahire, Rakesh Suryawanshi
Club Cycle: A Secure, Multi-Tiered Ecosystem for Campus Organization Management
Sharavana Ragav S, Aryan Sharma, Vibhu Sahu, Rajasekar Velswamy
Tamil Nadu 2026 Assembly Election Prediction Using Machine Learning and Dravidian Social Media Sentiment Analysis
Blesson Xavier M, Chirpparasan P, Hariganesh A, Kuduminathan P, Mrs. L. Shakira Banu, M.E
A Comprehensive Review and Recommendation Of AI Tools For Women Safety (SheShield)
Deepak Tyagi, Faiz Khan, Rooban Agarwal, Satish Kumar Soni, Uruj Jaleel
Autonomous Multi-Agent Pipeline for Biomedical Research Automation
Murugeswari K, Pradeesh S, Syed Umar Nafeez G, Vimal M, Vinny Sam Francis V
LARGE SCALE DATA AUDIT THROUGH BIG DATA TECHNOLOGIES
Anbarasi N, Prasanna P, Praveen Kumar M, Surendar N D, Venkadesh K
An Adaptive Stream-Native Anomaly Detection Framework Using Hybrid Unsupervised Learning
Jeevalakhmi K, Nithish T, Prathap S, Ramkumar K, Vijay M
SMART EMERGENCY VEHICLE ALERT SYSTEM USING ML & IoT
Parimala M, Nijanthan M, Shiam Alex S, Sibivarma S, Sugumaran
CYBERBULLYING DETECTION SYSTEM USING MACHINE LEARNING
Hovarthan S, Kishohar S, Mohamed Hathil M, Mugunthan R
The Evolution and Growth of IT Sector in India
Aditi Chaudhary, Anshul Sharma, Chitwan Ahlawat, Abhay Dhiman, Abhay Verma, Usha Kumari
AI-POWERED VIRTUAL MEETING SUMMARIZER
Shreyas Chaudhari, Aditya Ghamat, Vivek Billa, Chirag Lokhande, Dr. Ramesh Shahabade
CAR RENTAL BOOKING SYSTEM USING MERN STACK
Pooja Karche, Namrata Chaple, Komal Kapare, Pratik Shirsath, Prof. S. S Bere and Prof. P. S. Jagtap
AI-Augmented Business Intelligence Framework for Predictive and Prescriptive Analytics
Asharaf Pulikkalakath, Prasad J C
EXPLAINABLE AI BASED ARRHYTMIA MONITORING SYSTEM
PARIMALA M, VIGNESHWAR P, SAYOOJ KUMAR VS, SABARI KANNAN R, VISHWA RC
CampusXChange: A Django-Based Smart Student Marketplace Platform for Campus-Specific Peer-to-Peer Commerce
Siddesh Nilak, Omkar Pandit, Nihalveer Singh, Sumit Patil, Avinash Sonule
PCOD/PCOS HEALTH TRACKER: A SMART SYSTEM FOR MONITORING WOMEN’S HEALTH
Bhagyashree Kajale, Rameshwari Kamble, Archita Khedekar, V.A Bhamre
Digital Complaint Box System
Jayant Vijay Bhagat, Sankalp Santosh Bhosle, Dipak Ramesh Chavan, Prof. P. S. Shekhar
CRIME VIDEO ANALYSIS AND SUMMARIZATION DASHBOARD
Jeevalakshmi K, Santhosh J, Sivabalan M, Srikaran R, Sriram R
ComplaintHub: A Full-Stack Smart Complaint Management System with Real-Time Tracking and Intelligent Resolution Framework
Rishikesh Vedpathak, Aiman Sayyed, Rudra Tikekar, Sahil Tribhuvan, Prof. R. C. Suryawanshi
Smart Sanitation Complaint System Using QR Code,GPS Tracking, and Real-Time Image Verification
Ayush Sarvagod, Aniket Tavar, Jishan Pathan, Prof. Kanchanmala More
IoT-Based Water Quality Monitoring System
Amogha Nair, Hitesha Patil, Lavanya Sawant, Adiba Sayyed
EDGE AI ENABLED SMART MEDICAL DISPENSING AND PATIENT CARE ROBOT FOR HOSPITAL ROOMS
DR. AARTHY R, NAGALAKSHMI B, NEHA S KUMAR, SUKITHA S
Sign Language Recognition Using Deep Learning-A Review
Karthik Reddy A, Kushal Reddy K, Mallikarjuna S, P Akash Patil, Dr. Muhibur Rahaman T.R
Women Safety Alert System: An Android Application for Real-Time Emergency Response
Sakshi Singh, Trupti Sanap, Mahika Singh, Kalpesh Patil
DeepScan: A Heuristic-Based Framework for Deepfake and AI-Generated Image Detection Without Neural Network Inference
Neelesh N Shrinivasan, M. Sreedharan, Sriramji P, H. Mary Shiny
“AIVA: An AI-Powered Multi-Functional Virtual Assistant Mobile Application Using Flutter”
Komal Gulve, Kirti Nazirkar, Sanika Veer, Mohini Narwade, Mr. D.S. Jaybhay
Sentiment Analysis of Comments Received Through E- Consultation Module Software
Karuturi Shravani, Kavana K, N Suchitra, Preethi K, Dr.Muhibur Rahman T.R
Optimized Ensemble Intrusion Detection: Balancing Data with SMOTE-ENN and Feature Selection via Jaya Algorithm
Subba Reddy K, Nikhitha Dhayepule
Blockchain-Based Anti-Counterfeit Product Identification System: A Comprehensive Survey
Alam Basha N, Avinash V K, B K Raghavendra, Dandiya Mohammad Kaif, Dr. Muhibur Rahman T R
ConformityGate: Intelligent Honeypot-Based Intrusion Detection Framework for Secure IoT Networks
Deepak Marlecha, Zarana J Shah, Rishab P K, Netrang V Davey, Aditi Singh, Ms Charulatha R T
AI-Based Smart Forest Safety Monitoring System Using Real-Time Risk Analysis
N. Shahinaz, Nanditha N M, Netravathi Hosamani, K.V. Hima Rashmi, Dr. Muhibur Rahman T.R
Dormant Asset Rental for Small Businesses: A Web-Based Digital Marketplace for SME Asset Monetisation
Lalit Sabale, Chaitanya Panmand, Somnath Mulik, Rishabh Pawar, Dr. A R Sonule, Dr. Manoj M Deshpande, Dr. A S Deore
IOT BASED HEALTH CARE PREDICTION USING AI
Rajadurai N, Nagajothi P, Rajalakshmi K, Sridevi M, Yobhashini D
HistoAssist: A Production-Ready Full-Stack AI Framework Bridging Deep Learning Histopathology and Empathetic Patient Communication
Ali Khan Ayyub Khan, Altaf Ahmed Kasu, Khan Umair Abdul Salam, Md Yusuf Ansari, Alfiya Mulla, Zeeshan Khan
CYCLE CARE: A FLUTTER-BASED MENSTRUAL CYCLE TRACKING AND HEALTH MANAGEMENT APPLICATION
P.S Jogdand, S.S Gharat, H.R Deshmukh, S.V Ankam
Solutions of Burgers' Equation for Modeling Pulsatile Blood Flow in Arteries using Homotopy Analysis Method
Dr. Manoj Yadav*, Prof. Diwari Lal
Study Material Sharing Platform
Atharva A. Bhosale, Dewank B. Badiger, Sameer Bisen, Adityaraj Godbole, Smita Chunamari
A Digital Solution to Wrong E-Challan
Suyash S. Salunkhe, Vedika S. Sakharkar, Aditya D. Sharma, Adwait R. Velankar, Rakesh C. Suryawanshi
AEGIS-X: AI-POWERED AUTONOMOUS SURVEILLANCE AND COMMAND SYSTEM
SHAKIRA BANU L, ME., PARVAZE AHAMED, PRABHU C, SRIMURUGAN B, SHAGUL HAMMED J M
Job Description and Resume Matcher
Harsh Pandurang Korde, Soham Hanumant Shinde, Sumit Rajdev Mourya, Kanchan More
Machine Learning in Career Assistance and Job Application Automation: A Comprehensive Review
Seema, Sanjana R Bharade, Eshwari, Ananya
Multi-Agent Orchestration for Enhanced Text-to-SQL Generation: A Schema-Aware Approach with Self-Correction
Vijay M. Rakhade, Parth Gosavi, Chirag Kotkar, Pranav Kashmire, Siddharth Tripathi
A Review of Optical Character Recognition for Handwritten Hindi Text
Divya D, Bhuvaneshwari V, Shiva Kumar Swamy J, Siddu, Muhibur Rahman T.R, Anita Patil, Dadapeer
Phishing Website Detection using URL-Based Machine Learning for Real-Time Browser Security
Mohammed Zunaid, Puneeth MP, P Abhishek, Rishi Kumar P, Dr. Muhibur Rahaman T.R
A Secure and Intelligent Drone-Based Healthcare Logistics System with Real-Time IoT Monitoring
Dr. V. Ponniyinselvan, R. Srihariharan, S. SuryaPrakash, S. Poovarasan, K. Sandhuru
Agro Vision AI: A Web-Based Intelligent Agricultural E-Commerce Platform
Gopal Patel, Dhannajay Gupta, Kamal Kumar, Dasharath Gupta, Kumar Bibhuti Bhushan
Deep Learning Based Facial Emotion Recognition & Intelligent Facial Affect Detection
Anubhav Sharma, Kajal Muania, Muskan Sen, Tanu Pal, Ujjwal Goel, Mr. Rooban Agrawal, Dr. Uruj Jaleel, Dr. Satish Soni
AI-Based Maternal Healthcare Monitoring: A Comprehensive Survey
Adithi R, Akanksh R, Harshitha T, Shaik Khaja, Muhibur Rahman T.R, Anita Patil
Postpartum Depression Risk Prediction Model: Leveraging Machine Learning for Early Detection and Preventive Care
Tanish Shingade, Abhinav Ragam, Shriddhar Patil, Nikhil Nikam, Prof. Smita Chunamari
A MERN-Stack Blog Application with Gemini-3-Flash Integration
Aman Narayan, Abhinav Nigam, Abhishek Jaiswal, Suraj Kushwaha
“A Study on the Impact of Digital Banking Adoption on Customer Satisfaction and Financial Performance in Pune”
Atharva Somnath Kadam, Laxman Kalke, Dr. Pradyuman Shastri
Spatially-Gated CNN-Transformer Hybrids for Pneumonia Classification: A Unified Framework for Metric-Optimized Local-Global Explainability
Ch Mydhili*, B Madhav Rao
Automated Misinformation Detection in Online News Using NLP and LSTM-Based Deep Learning
M Sasidhar*, M Krishna
MACHINE LEARNING BASED TRAVEL MANAGEMENT AND OPTIMIZATION SYSTEM
Aditya Kumar Yadav, Rohit Yadav, Nishant Singh, Shivam Verma, Ms. Deepika Pandey
DEVELOPMENT OF A CONVOLUTIONAL NEURAL NETWORK FOR BINARY IMAGE CLASSIFICATION OF BEARS AND PANDAS WITH APPLICATIONS IN DIGITAL IMAGE PROCESSING
P Aarti Pai, Khushi, Sudarshan SR, H Mary Shyni
Revolutionizing Precision Agriculture with Machine Learning: Current Progress and Future Directions
Tajbir Singh, Satinder Kaur, Satveer Kour, Sandeep Kaur
Artificial Intelligence: Revolutionizing the Future of Medicine: Intelligent Systems, Deep Learning, and the Next Generation of Clinical Care
Nitin kumar, Jatin kumar, Shivam, Kanika Prajapati, Rooban Aggarwal, Satish kumar soni, Uruj jaleel
Design of an Integrated Model for Neuro Symbolic GenAI and RAG Driven Personalized Cardiometabolic Care Sets
Sagar Fernandes, Dr. Sangeeta Vhatkar
INTELLION BOT AI: FUTURE - READY INTELLIGENT BOT SYSTEM
Laxmi yadav, Harshita Singh, Anshika Yadav
CrackIt: An AI-Powered Interview Preparation and Performance Analysis Platform
Kartik Saini, Shivam Joshi, Mayank Sharma, Khushal Singh, Vishal Singh
Smart Agriculture System Using Internet of Things
Nikita Chauhan, Mohd. Sarim, Krishan kumar, Harshita, Khushi, Usha Kumari, Satish kumar soni, Uruj Jaleel
A SYSTEMATIC REVIEW OF AI POWERED FOOD RECOGNITION SYSTEM
Harsh Pachauri, Megha Rajput, Krishna Kaushik, Dharmesh Rawat, Pankaj Saraswat, Alok Singh Jadaun
Research Design Approaches in Cybersecurity Studies: A Comprehensive Review of Methods, Challenges, and Future Directions
Kamsali Aishwarya, Keerthi h, Keerthi somaraddi, M Charan, Dr. Muhibur Rahman T.R
Data Visualization Techniques and Actionable Insights
Ansh Tyagi, Arpit Tyagi, Aditya Adhana, Ms. Tanya Jain, Dr. Uruj Jaleel, Dr. Satish Kumar Soni
NeoAI: Socializing Reinvented with AI
Rampal Gupta, Ms. Deepika Pandey
AI-Powered Digital Justice System: A Proposed Smart Platform for Complaint Filing and Tracking – A Review
Chilukuri Venugopal, B N Gowtham, Basava Prabhu, Chandrasekhar H, Asst. Prof. Nagateja P
A Comprehensive Study on Privacy-Preserving Machine Learning (PPML) in Modern AI Systems
Naveen Kumar, Shiva Kumar, Paras Kaushik, Ms. Usha Kumari, Mr. Satish Kumar Soni, Mr. Uruj Jaleel
SMART EXPENSE FINANCE REPORTING USING AI
Gufran Siddiqui, Mohammad Arif, Saurabh Mishra, MD Arif, Mr. Bibhuti Kumar
SafeRoute AI: A Comprehensive Review of Safety-Aware Intelligent Navigation Systems
E Harsha, K P Sai Pravallika, Mahima Swaroopa C K, Deekshitha B, Muhibur Rahman T R
Zero Hunger: Smart Food Distribution Platform
Sharana Kumar K, Ajay Kumar Gouda S, Vaishnavi Y, MD.Akif Bari Shaik, Muhibur Rahman T.R, Anita Patil
Can everything be Solar Powered? – A Comprehensive Analysis
B.md. Saif, Prashanth Patil, Rahul Naik, Harshavardhan S, Dr. Muhibur Rahaman T.R
EXPENSE MATE
Shubham, Aniket Gupta, Manish Patel, Abhishek Yadav, Mrs. Prachi Yadav
Smart Healthcare System for Automated Disease Detection and Cure Recommendation through Medical Imaging
Y.G. Sanjana, Shravan Kumar. V, T. Manikanta, Vaishnavi. J, Anita S Patil
ArogyaAI: An Integrated LLM-Powered Healthcare Intelligence System
Chetan, Harshit Kumar, Atul Tyagi, Anshul Rastogi, Divyanshi Muvaniya, Dr. Brijesh Kr. Gupta
Customer Churn Analysis Using Python and Data Analytics for Telecom Industry
Kartik Bhargav, Khushi, Lalit Kumar, Mayank Bhardwaj, Dr. Brijesh Kumar Gupta
Survey-Based Research in Software Engineering
Kavya K, Layashree L, Raziya M, S Vaishnavi, Dr. Muhibur Rahman T.R
FinGenie: Design and Implementation of an AI-Assisted Financial Literacy Platform for Adolescents
Sowmya Prabhu, Soumya Sree T., Roopa Sai T., Venkat Ravi, Muhibur Rahman T.R, Anita Patil
Classification and Analysis of Disease over Symptoms using AI
Peetambar, Amir Ansari, Samrat Kartikey Maurya, Mr. Dileep Kumar Gupta
Ethical and Legal Implications of Using Generative AI in Academic Research
M S Saniya Khouser, Meghana D, Meghana Y, Mounika Y R, Dr. Muhibur Rahman T.R
MedRAG Nexus: An AI-Powered Health Intelligence System Using Retrieval-Augmented Generation and Agentic AI
Sandeep Tomar, Abhishek Soam, Shekhar Tomar, Tanya Chaudhary, Sandhya Kashyap, Dr. Brijesh Kr. Gupta
Full-Stack Hospital Management System
Shivraj, Varun Sirohi, Saurabh Kumar Nirwan, Shivansh Chaudhary, Tanya Jain, Dr.Uruj Jaleel, Dr. Satish Kumar Soni
AI Based Food Donation and Distribution System
B Bhargavi, Chandana B A, Kasireddy Rithu Reddy, V R Kalyani, Anita Patil
AI-Based Mental Health Chatbot using LangChain and LangGraph
Sachin saini, Yashika Rathi, Yash Kumar, Shweta Rani, Dr Brijesh kr. Gupta
Enhanced Distributed Flood Detection and Alert System Using Deep Learning
NageswaraRao Sirisala, Srinivasulu Sirisala, Anitha Yarava
Deep Learning Framework for Prior Identification of Threats in Social Media Interactions
NageswaraRao Sirisala, Srinivasulu Sirisala, Anitha Yarava
A Study on Multi-Location Ledger Using DPOS (DP46) in a FPL Hyundai Automobile Dealership
Kumar. Nandini, Dr. Kumarakannan.R
A STUDY ON THE ANALYSIS OF POST-PURCHASE EXPERIENCE USING POWER BI WITH REFERENCE TO SOLSTROM ENERGY SOLUTION PVT LTD
Mr. S. Balakrishnan, Dr. K. Kokila
A STUDY ON IMPACT OF SOCIAL MEDIA MARKETING ON CONSUMER BUYING BEHAVIOUR - DIGITAL MARKETING SECTOR, CHENNAI.
Ms. K. Thrisha, Dr. K. Kokila
Abstract
DEVELOPMENT OF A CHILD ABUSE MONITORING AND REPORTING SYSTEM USING A NATURAL LANGUAGE PROCESSING MODEL
Chibuike Eusebius Nnaemeka, Obikwelu Raphael Okonkwo, Njideka Nkemdilim Mbeledogu
DOI: 10.17148/IJARCCE.2026.15401
Abstract: Child abuse remains a major global concern that threatens the physical, emotional, and psychological development of children. Despite growing awareness and legal frameworks aimed at protecting children, many abuse cases remain undetected or are reported too late due to inadequate monitoring systems, poor data quality, fear of stigma, and ineffective reporting mechanisms. Psychological abuse is often difficult to detect because it does not leave visible physical evidence and is therefore frequently underreported. In addition, the rapid expansion of digital communication platforms has created new avenues through which abusive behaviors such as cyberbullying, harassment, and threats can occur, making traditional monitoring approaches insufficient. This study presents the development of the Child Abuse Monitoring and Reporting (CAMR) System that uses Natural Language Processing (NLP) to enhance the detection, monitoring, and reporting of child abuse cases. The proposed system employs sentiment analysis and Named Entity Recognition (NER) techniques to identify emotional tones and relevant entities in text that may indicate abusive interactions. The system is implemented using the Naïve Bayes algorithm to classify text and detect potentially abusive content.
The system's performance is evaluated using accuracy, precision, recall, and F1-score, along with User Acceptance Testing (UAT), to assess its effectiveness and usability. The study demonstrates that integrating NLP techniques into child protection systems can enhance early detection of abuse, enable automated monitoring of large volumes of textual data, and support timely intervention. The proposed system contributes to improving child protection strategies by providing a scalable and efficient technological solution for monitoring and reporting abuse, particularly within digital communication environments. Ultimately, the system supports government agencies, institutions, and child protection organizations in safeguarding children and responding more effectively to abuse cases.
Keywords: Child Abuse; Child Protection; Abuse Detection; Psychological Abuse; Cyberbullying; Harassment; Threats.
The system's performance is evaluated using accuracy, precision, recall, and F1-score, along with User Acceptance Testing (UAT), to assess its effectiveness and usability. The study demonstrates that integrating NLP techniques into child protection systems can enhance early detection of abuse, enable automated monitoring of large volumes of textual data, and support timely intervention. The proposed system contributes to improving child protection strategies by providing a scalable and efficient technological solution for monitoring and reporting abuse, particularly within digital communication environments. Ultimately, the system supports government agencies, institutions, and child protection organizations in safeguarding children and responding more effectively to abuse cases.
Keywords: Child Abuse; Child Protection; Abuse Detection; Psychological Abuse; Cyberbullying; Harassment; Threats.
Abstract
Real-Time Explainable Malware Detection with Automated Response
Mrs Ayesha Azeeza, Dr Nasreen Taj M
DOI: 10.17148/IJARCCE.2026.15402
Abstract: Modern computing systems are constantly exposed to unknown and evolving threats, making it difficult to ensure reliable protection using traditional security methods alone. While machine learning-based approaches have improved the ability to detect malicious activities, many of these systems still fail to clearly explain their decisions or respond quickly enough when a threat is identified. As a result, there is often a gap between detection, understanding, and action.
This paper presents a real-time malware detection framework that focuses on explainability, traceability, and automated response. The proposed system monitors system-level behavior and analyzes process activities using a transformer-based model that captures patterns over time. When a process is identified as suspicious, the system provides a clear, humanreadable explanation describing why it is considered malicious, along with traceable details such as where the activity originated and how it progressed within the system.
To minimize the impact of potential threats, the framework includes an automated response mechanism. If a process exceeds a defined risk threshold based on abnormal behavior, it is immediately terminated or isolated. In addition, a structured report is generated and stored, allowing users or analysts to review the complete details of the event whenever required.
Unlike existing approaches that treat detection and response separately, this work integrates detection, explanation, and action into a single unified framework. This not only reduces response time but also improves the clarity and usability of the system, making it more practical for real-world cybersecurity scenarios.
Furthermore, the system is designed to operate in real time without introducing significant overhead, ensuring that it remains efficient even in dynamic environments. By combining accurate detection with clear explanation and immediate response, the proposed approach aims to improve both trust and effectiveness in modern malware defense systems.
Keywords: Malware Detection, Explainable AI, Behavioral Analysis, Transformer Models, Real-Time Monitoring, Automated Response, Traceability, Cybersecurity
This paper presents a real-time malware detection framework that focuses on explainability, traceability, and automated response. The proposed system monitors system-level behavior and analyzes process activities using a transformer-based model that captures patterns over time. When a process is identified as suspicious, the system provides a clear, humanreadable explanation describing why it is considered malicious, along with traceable details such as where the activity originated and how it progressed within the system.
To minimize the impact of potential threats, the framework includes an automated response mechanism. If a process exceeds a defined risk threshold based on abnormal behavior, it is immediately terminated or isolated. In addition, a structured report is generated and stored, allowing users or analysts to review the complete details of the event whenever required.
Unlike existing approaches that treat detection and response separately, this work integrates detection, explanation, and action into a single unified framework. This not only reduces response time but also improves the clarity and usability of the system, making it more practical for real-world cybersecurity scenarios.
Furthermore, the system is designed to operate in real time without introducing significant overhead, ensuring that it remains efficient even in dynamic environments. By combining accurate detection with clear explanation and immediate response, the proposed approach aims to improve both trust and effectiveness in modern malware defense systems.
Keywords: Malware Detection, Explainable AI, Behavioral Analysis, Transformer Models, Real-Time Monitoring, Automated Response, Traceability, Cybersecurity
Abstract
SOUNDFOREST: MONITORING FOREST BIODIVERSITY USING AI-POWERED SOUND CLASSIFICATION
MATHIR VISHNU S, REVATHI A
DOI: 10.17148/IJARCCE.2026.15403
Abstract: For ecological preservation and environmental management, monitoring biodiversity in forest ecosystems is essential. Conventional monitoring techniques, like camera traps and direct observation, are frequently labor-intensive, time-consuming, and have a limited geographic and temporal reach. In this paper, we propose SoundForest, an AI-powered system that analyse natural ambient audio recordings to classify and track forest biodiversity. The system continuously gathers soundscape data from forest environments using edge or remote sensors. After preprocessing to eliminate background noise, the audio data is converted into time-frequency representations like Mel-Frequency Cepstral Coefficients (MFCCs) and Mel spectrograms. Deep learning models, namely Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Audio Spectrogram Transformers (AST) for multi-class sound classification, are trained using these representations. The models can detect changes in acoustic biodiversity over time and recognize vocalizations specific to a species. Anthropogenic sounds, such as chainsaw activity, gunshots, car engines, and human voices, are also recognized by SoundForest in order to detect wildlife and warn of unlawful logging, poaching, or unauthorized entry into protected forest areas. Even in remote locations, real-time, decentralized monitoring is made possible by the system’s support for deployment on low-power edge devices like Raspberry Pi. Supported by the system are heatmaps of biodiversity, behavioural patterns, and species distributions.
Keywords: Biodiversity Monitoring, Acoustic Sensing, Environmental Sound Classification, Deep Learning, CNN, RNN, Audio Spectrogram Transformer, Wildlife Conservation
Keywords: Biodiversity Monitoring, Acoustic Sensing, Environmental Sound Classification, Deep Learning, CNN, RNN, Audio Spectrogram Transformer, Wildlife Conservation
Abstract
Predicting Customer Churn Using Advanced Machine Learning Ensemble Methods with Sentiment Analysis Integration
Sakthi Dharan S, Dr. A. Revathi
DOI: 10.17148/IJARCCE.2026.15404
Abstract: Customer churn represents a critical business challenge in telecommunications and subscription-based services, with annual revenue losses exceeding billions of dollars globally. This research presents a comprehensive machine learning framework for predicting customer churn using an ensemble of five advanced algorithms: Logistic Regression, Ran-dom Forest, XGBoost, LightGBM, and Gradient Boosting. We integrate structured behavioral data with sentiment analysis derived from customer feedback using TextBlob, incorporating dynamic sentiment weighting that adjusts churn probabilities by up to 45%. The framework employs SMOTE (Synthetic Minority Over-sampling Technique) for handling class imbalance and SHAP (SHapley Additive Explanations) for model interpretability. Experimental results on a telecommunications dataset of 7,043 customers demonstrate that the XGBoost classifier achieves supe-rior performance with 89.4% accuracy, 0.86 F1-score, and 0.95 AUC-ROC, outperforming baseline models by 7.8%. The sentiment weighting mechanism reduces false negatives by 23%, significantly improving identification of at-risk customers. The complete system is deployed as an interactive Gradio web application with real-time sentiment analysis, enabling businesses to make data-driven retention decisions. This research contributes a production-ready, interpretable churn prediction system that integrates multiple ensemble methods with sentiment-based probability adjustment for enhanced predictive accuracy.
Keywords: Customer Churn Prediction, Ensemble Learning, XGBoost, LightGBM, Random Forest, SMOTE, Senti- ment Analysis, SHAP, Gradio Deployment
Keywords: Customer Churn Prediction, Ensemble Learning, XGBoost, LightGBM, Random Forest, SMOTE, Senti- ment Analysis, SHAP, Gradio Deployment
Abstract
AI INTERVIEW PREPARATION SYSTEM
Harish Mythrayan T, Ashwin Shano S A, Baranidharan R
DOI: 10.17148/IJARCCE.2026.15405
Abstract: This paper introduces an AI-driven Interview Preparation System, which aims to improve the candidate preparation to improve recruitment workflows by intelligent automation and multimodal analysis. The system incorporates a progressive, multi-level interview engine that enables the injection of context-aware questions using cutting-edge language models to allow the candidate to experience various structured interview situations that range from the basic HR interaction to a technical-level expert. To better the quality of assessments, the platform uses video- based real-time analysis of emotions and audio for response analysis to offer end-to-end evaluation on multiple dimensions such as relevance, clarity, confidence, and technical accuracy.
The architecture designed relies on a modular architecture comprising user interface using web and high-performance backend framework to handle the scalable processing. The frontend offers country-specific role-based dashboards for candidates, HR professionals, and administrators with various supporting features, such as scheduling interviews, tracking performance, and managing users. The backend makes use of the AI services for question generator and answer evaluation services whereas, on the other hand, facial emotion analysis models allows for better behavioural insights during the interview sessions. Additionally, semantic matching methods are also used to match candidate profiles to job requirements, making it possible to make better hiring decisions.
Experimental observations show that the system simulates realistic interview situations and at the same time offers usable feedback in form of detailed reports of the performance. The merging of multimodal information sources adds great value in evaluation depth that is not found within the text-based systems. Furthermore, the platform guarantees flexibility by configurable parameters such as levels of the interviews, thresholds for scoring and user roles, making it adaptable to different recruitment cases. The system plays a role in filling the gap between traditional methods of interview preparation along with modern artificial intelligence solutions by providing an end-to-end intelligent interview ecosystem.
Keywords: Artifical intelligence, interview preparation system, multimodal analysis, emotion recognition, natural language processing, candidate evaluation, machine learning, human resource management, semantic matching, fastapi.
The architecture designed relies on a modular architecture comprising user interface using web and high-performance backend framework to handle the scalable processing. The frontend offers country-specific role-based dashboards for candidates, HR professionals, and administrators with various supporting features, such as scheduling interviews, tracking performance, and managing users. The backend makes use of the AI services for question generator and answer evaluation services whereas, on the other hand, facial emotion analysis models allows for better behavioural insights during the interview sessions. Additionally, semantic matching methods are also used to match candidate profiles to job requirements, making it possible to make better hiring decisions.
Experimental observations show that the system simulates realistic interview situations and at the same time offers usable feedback in form of detailed reports of the performance. The merging of multimodal information sources adds great value in evaluation depth that is not found within the text-based systems. Furthermore, the platform guarantees flexibility by configurable parameters such as levels of the interviews, thresholds for scoring and user roles, making it adaptable to different recruitment cases. The system plays a role in filling the gap between traditional methods of interview preparation along with modern artificial intelligence solutions by providing an end-to-end intelligent interview ecosystem.
Keywords: Artifical intelligence, interview preparation system, multimodal analysis, emotion recognition, natural language processing, candidate evaluation, machine learning, human resource management, semantic matching, fastapi.
Abstract
A Smart Blockchain-Based Platform for Fake Product Detection and Secure Product Verification
Khan Fardeen, Adnan Ansari, Khan Abusayma, Mansi Shimpi
DOI: 10.17148/IJARCCE.2026.15406
Abstract: Users in today’s digital marketplace often struggle to verify the authenticity of products due to the rapid growth of counterfeit goods and the lack of reliable verification systems. Most existing product authentication methods rely on centralized databases or manual verification processes, which are vulnerable to data manipulation and lack transparency. These systems do not effectively provide secure traceability, product verification, and consumer trust within a single platform. This paper proposes a blockchain-based fake product detection system that utilizes decentralized technology to improve product authentication and supply chain transparency. The system records product information such as manufacturing details, batch number, and ownership history on a blockchain network, ensuring that the data remains secure and immutable. Consumers can verify the authenticity of products by scanning a QR code or unique product identifier linked to the blockchain record. In addition, the system enables stakeholders to track product movement across different stages of the supply chain. This helps in preventing counterfeit distribution and allows reliable product verification. Overall, the design demonstrates how a transparent and secure blockchain-based system can support better product authentication, consumer trust, and protection against counterfeit goods.
Keywords: Blockchain technology, Fake product detection, Product authentication, Counterfeit prevention, Supply chain transparency, QR code verification, Decentralized ledger system, Product traceability, Consumer trust, Secure product verification.
Keywords: Blockchain technology, Fake product detection, Product authentication, Counterfeit prevention, Supply chain transparency, QR code verification, Decentralized ledger system, Product traceability, Consumer trust, Secure product verification.
Abstract
SAFE DRIVE IOT: Intelligent System to Prevent Drunk Driving
Prof. Liril George, Mihir Kishor Kumbhar, Ravitej Hiremath, Soham Arun Chandorkar
DOI: 10.17148/IJARCCE.2026.15407
Abstract: This project introduces Safe Drive IoT, a crucial vehicle security and safety system engineered to directly confront drunk driving, a leading cause of road fatalities. The core design integrates an IoT framework with essential hardware components: an alcohol detection sensor, a GPS module, a GSM communication module, and an ignition locking mechanism. The system's operation is straightforward and uncompromising: upon vehicle entry, the alcohol sensor monitors the driver's breath. If the detected alcohol concentration surpasses the safety limit, the system immediately triggers a warning via an integrated buzzer and, most importantly, activates the ignition lock, preventing the vehicle from starting.
In the event of an attempted start under intoxicated conditions, the system utilizes the GPS module to determine the vehicle's precise location and the GSM module to transmit an instant, geolocated alert (SMS) to emergency contacts and enforcement agencies. Safe Drive IoT provides a robust, real-time safety layer that effectively enforces sobriety behind the wheel, ensuring that a simple, yet life-saving, mechanism governs vehicle operation.
Keywords: Drunk driving, IoT, alcohol detection, GSM, MQ-3, GPS.
In the event of an attempted start under intoxicated conditions, the system utilizes the GPS module to determine the vehicle's precise location and the GSM module to transmit an instant, geolocated alert (SMS) to emergency contacts and enforcement agencies. Safe Drive IoT provides a robust, real-time safety layer that effectively enforces sobriety behind the wheel, ensuring that a simple, yet life-saving, mechanism governs vehicle operation.
Keywords: Drunk driving, IoT, alcohol detection, GSM, MQ-3, GPS.
Abstract
Smart Waste Segregation and Monitoring System Using IoT
Prof. Rita Pawade, Vipul S. Manbhe, Shreya S. Wasnik, Kashish R. Shende, Tushar V. Gore, Govinda K Belekar, Yuvraj L. Wadaskar
DOI: 10.17148/IJARCCE.2026.15408
Abstract: The increasing generation of waste demands efficient and accurate segregation methods. This project presents an IoT-based Automated Waste Segregation System that classifies waste into four categories: wet, dry, plastic, and metal. The system uses an ESP32 microcontroller integrated with moisture, inductive proximity, infrared, and ultrasonic sensors for detection and monitoring.
A conveyor belt mechanism with stepper and servo motors enables automatic sorting based on multi-stage sensing. The system also uses MQTT and HiveQL cloud for real-time monitoring through a web dashboard. Key advantages include low cost, reduced human effort, improved accuracy, and efficient handling of biodegradable wet waste.
This solution supports smart waste management and contributes to better recycling and environmental sustainability.
A conveyor belt mechanism with stepper and servo motors enables automatic sorting based on multi-stage sensing. The system also uses MQTT and HiveQL cloud for real-time monitoring through a web dashboard. Key advantages include low cost, reduced human effort, improved accuracy, and efficient handling of biodegradable wet waste.
This solution supports smart waste management and contributes to better recycling and environmental sustainability.
Abstract
MediCard+ An AI-Driven Smart Healthcare System for Centralized Medical Records and Decision Support
Prof. Manoj Babar, Sonali Patil, Purva Powar Patil, Yash Patil, Aneesh Patil, Pranita Yadav
DOI: 10.17148/IJARCCE.2026.15409
Abstract: This paper presents MediCard+, an AI-driven centralized healthcare system designed to address fragmented medical records and inefficient clinical decision-making. The platform enables secure storage, retrieval, and management of patient records across multiple hospitals using role-based access control. A key feature is an AI- powered chatbot that uses Natural Language Processing (NLP) and a pre-trained LLaMA-based model to analyze medical data and provide real-time insights. It supports medical history summarization, detection of critical conditions, and prediction of long-term disease risks. To ensure privacy, MediCard+ includes patient-controlled data sharing and a privacy-preserving AI pipeline that prevents direct exposure of sensitive data. The system uses MongoDB for structured data, cloud storage for documents, and a Node.js backend for scalability. Experimental results show improved data accessibility, faster response time, and better clinical decision support, making MediCard+ a scalable and secure solution for modern healthcare systems.
.
Keywords: Centralized EHR, Clinical Decision Support, Healthcare AI Chatbot, Role-Based Access Control, Medical Data Security, NLP in Healthcare
.
Keywords: Centralized EHR, Clinical Decision Support, Healthcare AI Chatbot, Role-Based Access Control, Medical Data Security, NLP in Healthcare
Abstract
Real-Time Human Emotion Recognition and Analysis using DeepFace and OpenCV
Shaikh Asif, Ayan Dawat, Sayyed Anas, Shaikh Mufeez, Shah Mohd Sharique
DOI: 10.17148/IJARCCE.2026.15410
Abstract: - Facial Emotion Recognition, or FER, is at the heart of affective computing these days. It isn’t just making tech feel more human—it’s changing how we watch for mental health swings and beef up security, too. There’s a lot of excitement about what FER can do, but using these systems outside the lab brings some real headaches. Deep learning eats up a lot of computing power, and life is messy—dim rooms, people looking away, turning their heads. All that can trip up even solid models.
That’s what this study digs into. We put together a real-time emotion detection system using DeepFace and OpenCV. Everything hinges on deep CNNs—they’re built to spot seven main facial expressions: happy, sad, angry, fear, surprise, disgust, and neutral. We wanted anyone to be able to use it, so we made one smart change: every video frame gets resized before being processed. That one step sped things up, cut down on processor load, and kept our accuracy high. In the end, the system hums along at 25 frames per second, handling the messiness of the real world without breaking a sweat.
One piece really pops out—statistical tracking. The system doesn’t just spot an emotion and call it a day. It tracks how often each emotion shows up and maps out the emotional changes as a session goes on. Thanks to dynamic data structures, you get more than just quick snapshots—you see the whole story, how emotions shift over time. Bottom line? Pairing pre-trained VGG-Face models with this straightforward setup gets great results—even when faces disappear for a moment. The system rolls with whatever comes its way and just keeps going. It’s a solid base for real- time emotion analysis, and now researchers can actually track mood swings as they happen, or build interfaces that react right when you need them to. It’s a good step toward machines that actually get what people are feeling.
Keywords: - Face Detection, Emotion Recognition, Deep Learning, OpenCV, Artificial Intelligence.
That’s what this study digs into. We put together a real-time emotion detection system using DeepFace and OpenCV. Everything hinges on deep CNNs—they’re built to spot seven main facial expressions: happy, sad, angry, fear, surprise, disgust, and neutral. We wanted anyone to be able to use it, so we made one smart change: every video frame gets resized before being processed. That one step sped things up, cut down on processor load, and kept our accuracy high. In the end, the system hums along at 25 frames per second, handling the messiness of the real world without breaking a sweat.
One piece really pops out—statistical tracking. The system doesn’t just spot an emotion and call it a day. It tracks how often each emotion shows up and maps out the emotional changes as a session goes on. Thanks to dynamic data structures, you get more than just quick snapshots—you see the whole story, how emotions shift over time. Bottom line? Pairing pre-trained VGG-Face models with this straightforward setup gets great results—even when faces disappear for a moment. The system rolls with whatever comes its way and just keeps going. It’s a solid base for real- time emotion analysis, and now researchers can actually track mood swings as they happen, or build interfaces that react right when you need them to. It’s a good step toward machines that actually get what people are feeling.
Keywords: - Face Detection, Emotion Recognition, Deep Learning, OpenCV, Artificial Intelligence.
Abstract
Brain Tumor Detection
Talha Tidhare, Chaudhary Ashfaque, Bagwan Affan, Shaikh Maaz, Imran Shahid Abdul Rasheed
DOI: 10.17148/IJARCCE.2026.15411
Abstract: This paper presents a brain tumor detection system designed to assist early diagnosis using a multi-model machine learning approach. The system integrates MRI image analysis using Convolutional Neural Network (CNN) and symptom-based prediction using Random Forest. It combines both medical imaging and clinical data to improve accuracy and reliability. The system is implemented as a web-based application that allows users to upload MRI images or enter symptoms for preliminary screening. It targets healthcare support by providing fast, accessible, and effective tumor detection.
Keywords: Brain Tumor Detection, Machine Learning, Deep Learning, Convolutional Neural Network, Random Forest, MRI, Medical Image Analysis, Web Application.
Keywords: Brain Tumor Detection, Machine Learning, Deep Learning, Convolutional Neural Network, Random Forest, MRI, Medical Image Analysis, Web Application.
Abstract
The Impact of Artificial Intelligence and Digitalization on the Workforce: A Skill-Biased Technological Change and Human Capital Perspective
Harsh Bhatt, Amit Kumar Sahu, Mrs Harshita Gaikwad
DOI: 10.17148/IJARCCE.2026.15412
Abstract: The increasing pace of diffusion of Artificial Intelligence (AI) and digitalization is radically re-organizing labour markets world-over, but micro-level empirical data on the interaction of automation exposure and organizational reskilling to influence employee job-security perceptions are limited. This paper is based on the Skill-Biased Technological Change (SBTC) theory and Human Capital Theory (HCT) and creates a mediated moderation model where organizational reskilling provision mediates and moderates the association between AI exposure and job-security concern. A total of 240 respondents in the manufacturing, logistics, retail, healthcare, and IT industries were used to collect primary data using a structured questionnaire. Pearson Chi-square, correlation analysis, descriptive statistics and multiple regression (OLS) were used. The chi-square test also ensured that there was no direct significant relationship between automation exposure and job-security concern (χ² = 3.28, df = 8, p = 0.916). But the regression analysis indicated that the reskilling provision has strong and negative predictive result of job-security concern (β = -0.34, p < 0.001) and that AI exposure interacting with reskilling has a strong attenuative effect on concern (β = -0.21, p = 0.018). The perception of positive automation had a strong influence on the optimism regarding the creation of new jobs (β = 0.47, p < 0.001). These results build on the SBTC model by showing that the exposure to automation is not the determinant of psychological employment outcomes but the organizational human-capital investment reaction. The research provides evidence to inform policymakers to build reskilling infrastructures and organization leaders to build human-AI partnerships.
Keywords: Artificial Intelligence; Digitalization; Workforce Transformation; Skill-Biased Technological Change; Human Capital Theory; Job-Security; Reskilling; Mediated Moderation; Regression Analysis
Keywords: Artificial Intelligence; Digitalization; Workforce Transformation; Skill-Biased Technological Change; Human Capital Theory; Job-Security; Reskilling; Mediated Moderation; Regression Analysis
Abstract
Detection and Prevention of Lung Diseases Using Convolutional Neural Networks: A Comprehensive Review
Pradnya Meshram, Tanaya Kattekola, Yashashree Sankade, Sakshi Kature, Hitesh Chatur
DOI: 10.17148/IJARCCE.2026.15413
Abstract: Respiratory diseases such as pneumonia, tuberculosis, chronic obstructive pulmonary disease, and lung cancer continue to pose a major threat to global public health. Accurate and early diagnosis remains a critical challenge due to inter-observer variability, radiologist fatigue, and limited access to expert interpretation, particularly in resource- constrained settings. Recent advances in deep learning, specifically convolutional neural networks (CNNs), have enabled automated analysis of chest radiographs and computed tomography scans with accuracy approaching and, in some cases, exceeding expert-level performance. This paper presents a detailed review of CNN-based methodologies for lung dis- ease detection and prevention. The review synthesizes theoretical foundations, learning paradigms, benchmark datasets, state-of-the-art architectures, performance metrics, and prevention-oriented applications such as risk stratification and opportunistic screening. Key challenges related to data bias, explainability, regulatory compliance, and real-world deployment are discussed, along with future research directions toward robust and ethically deployable AI-driven diagnostic systems.
Keywords: Convolutional Neural Networks, Lung Disease Detection, Chest X-ray, Deep Learning, Medical Imaging, Preventive Healthcare
Keywords: Convolutional Neural Networks, Lung Disease Detection, Chest X-ray, Deep Learning, Medical Imaging, Preventive Healthcare
Abstract
Evaluating Customer Loyalty Program: Starbucks India
Somya Prasad, Vora Smit
DOI: 10.17148/IJARCCE.2026.15414
Abstract: This paper provides an empirical analysis of customer loyalty programs as a relationship management tool in experience-based service context with special references to Starbucks India. Although loyalty programs have become commonplace in the cafe and quick service restaurant (QSR) industry, there is little empirical research on how the particular features of the programs are converted into customer loyalty results based on implicit psychological and relational processes. To fill this gap, Structural Equation Modeling (SEM) will be used to test the interrelationships of the attributes of the loyalty programs, the perceived value and customer satisfaction, brand trust, and customer loyalty. The research is based on the relationship marketing theory, customer equity theory, expectancy-disconfirmation theory, and social exchange theory; it provides a sequential mediation model that is proven by empirical research. The findings have shown that the perceived value (β = 0.54, p < 0.001) is affected by the attributes of loyalty programs (β = 0.54, p < 0.001) and leads to customer satisfaction (β = 0.49, p < 0.001). Growth of brand trust (β = 0.46, p < 0.001) follows as a result of customer satisfaction and customer loyalty (β = 0.52, p < 0.001) is a resultant outcome. The analysis also indicates the relationship between perceived value and customer loyalty to be mediated by customer satisfaction (indirect effect = 0.18, p < 0.01). The degree to which the hypothesized relationship is significant is statistically significant and the structural model has good fit as indicated by various goodness-of-fit measures (CFI = 0.93, TLI = 0.92, RMSEA = 0.06). The findings add to the research on the loyalty program in the emerging markets by showing that the loyalty of the premium cafe type brands is best achieved by using the sequential process of value creation, satisfaction and building of trust and not by the single transactional incentives. The research has provided theoretical contributions to relationship marketing and customer equity literature and has also provided managerial implications that can be used by managers to create value-based loyalty programs in competitive and experience-based service settings
Keywords: Customer loyalty programs • Loyalty program attributes • Perceived value • Customer satisfaction • Brand trust • Customer loyalty • Relationship marketing • Structural Equation Modeling (SEM) • Premium café brand • Emerging market • India • Experiential services • Customer equity • Service marketing • Mediation analysis
Keywords: Customer loyalty programs • Loyalty program attributes • Perceived value • Customer satisfaction • Brand trust • Customer loyalty • Relationship marketing • Structural Equation Modeling (SEM) • Premium café brand • Emerging market • India • Experiential services • Customer equity • Service marketing • Mediation analysis
Abstract
AI-ENABLED NETWORK MONITORING SOLUTION
DR.D. VIMAL KUMAR, DR.A. REVATHI, K. ABIRAMI
DOI: 10.17148/IJARCCE.2026.15415
Abstract: This project develops an AI-Enabled Network Monitoring System to improve network monitoring. As networks grow with more users and data, manual monitoring becomes difficult. Traditional systems use fixed rules and cannot detect new issues effectively. This project uses the Isolation Forest algorithm to detect abnormal network activities. It analyses data like bandwidth, packet flow, and latency to learn normal behaviour and find unusual patterns. The system works in real time and sends alerts when anomalies are detected. Overall, this project provides a simple and automated solution to improve network performance and security while reducing manual effort.
Keywords: AI Monitoring, Network Traffic Analysis, Anomaly Detection, Machine Learning, Network Security, insulation timber, Python.
Keywords: AI Monitoring, Network Traffic Analysis, Anomaly Detection, Machine Learning, Network Security, insulation timber, Python.
Abstract
PREDICTIVE ANALYTICS FOR WORKFORCE REDUCTION IN MULTINATIONAL CORPORATIONS: A LOGISTIC REGRESSION APPROACH
Dr. Vimal Kumar D, Mr. Arockiya John Prasanna X, Ms Hemalatha A
DOI: 10.17148/IJARCCE.2026.15416
Abstract: Corporate workforce restructuring has become one of the defining business challenges of our era. Over the past several years, multinational corporations across technology, finance, and manufacturing have carried out large-scale layoffs driven by a mix of macroeconomic headwinds, rapid automation, shrinking profit margins, and rising debt loads. Yet despite the frequency of these events, most organisations still lack any reliable early-warning system to flag layoff risk before decisions are locked in. This study introduces a data-driven approach to predicting corporate layoff events. Using a synthetic dataset of 1,500 firm-year observations spanning 2020 to 2026, we built a logistic regression model that draws on financial indicators, operational metrics, and macroeconomic data. The model was trained, validated through stratified cross-validation, and tested against a comprehensive set of classification metrics. We engineered several composite features including revenue decline flags, high-automation indicators, debt-to-equity thresholds, and an overall risk score that meaningfully improved predictive accuracy. We then applied the trained model to estimate layoff probabilities for ten prominent global companies in 2025–2026. The model achieved a ROC-AUC of around 0.82, with well-balanced precision and recall. Our aim is to offer a practical, reproducible methodology that organisations and policymakers can genuinely use for early workforce risk assessment.
Keywords: corporate layoffs, logistic regression, workforce analytics, predictive modelling, MNC risk assessment, automation adoption, financial indicators, ROC-AUC.
Keywords: corporate layoffs, logistic regression, workforce analytics, predictive modelling, MNC risk assessment, automation adoption, financial indicators, ROC-AUC.
Abstract
LegalBot: An AI-Based Legal Chatbot System for Accessible Legal Guidance
Bhoomi Singh, Rishabh Pandey, Bhoomi Gupta, Prarthita Lade, Mr. Kashif Shaikh
DOI: 10.17148/IJARCCE.2026.15418
Abstract: Access to legal information remains a significant challenge for a large portion of the general public due to complex legal terminology, high consultation costs, and limited availability of legal professionals. This paper presents LegalBot, an AI-powered legal chatbot system designed to bridge this gap by providing real-time, conversational legal guidance to users. Built on a modern full-stack architecture comprising Python, FastAPI, and MongoDB, LegalBot integrates an external AI language model to generate context-aware responses to user queries. The system supports user authentication, persistent chat history, and a clean web-based interface. This paper describes the system architecture, development methodology, implementation details, and evaluation results. A comparative analysis with existing legal chatbot systems is also provided. Future enhancements including multilingual support, voice assistance, and mobile application deployment are discussed.
Keywords: Legal chatbot, Artificial intelligence, FastAPI, MongoDB, Natural language processing, Legal assistance, Full-stack system
Keywords: Legal chatbot, Artificial intelligence, FastAPI, MongoDB, Natural language processing, Legal assistance, Full-stack system
Abstract
UPI Fraud Detection Using Hybrid Machine Learning Models with Explainable Risk Scoring and Real-Time Monitoring
Tarra Sekhar, G. Vijaya Kumar
DOI: 10.17148/IJARCCE.2026.15419
Abstract: Unified Payments Interface has become a major transaction channel in day-to-day digital payments, which makes transaction security an important technical concern. Fraud detection in such environments is difficult because suspicious transactions form only a very small portion of the total data, while legitimate user activity can vary widely in amount, timing, and transaction type. This work presents a hybrid machine learning framework for UPI fraud detection that combines a baseline Random Forest model with an improved XGBoost-based detection model, supported by imbalance handling, engineered transaction features, explainable prediction analysis, and a real-time monitoring dashboard. The system is designed as a complete end-to-end pipeline consisting of dataset preparation, preprocessing, feature transformation, model training, fraud scoring, API-ready prediction flow, and interactive visualization. In the implementation, transaction attributes are transformed into a compact feature set that includes temporal behaviour, transaction-category indicators, and fraud-rate information by transaction type. Synthetic Minority Over-sampling Technique is used to reduce the effect of class imbalance during training. The trained model produces a fraud probability score, and a decision threshold of 0.6 is used for final classification. To improve transparency, SHAP-based feature explanations are integrated so that important factors behind a prediction can be viewed instead of treating the model as a complete black box. A Streamlit dashboard was developed to support manual fraud checking, risk-gauge visualization, live transaction simulation, feature-importance display, and monitoring analytics. Experimental results show strong discrimination between legitimate and suspicious transactions. From the recorded confusion matrix, the system correctly identifies 314224 legitimate transactions and 168 fraudulent transactions, with zero false positives and 181 false negatives. These results indicate that the proposed framework is highly reliable for safe-transaction confirmation and practically useful for risk-aware UPI monitoring applications. The work demonstrates that combining ensemble learning, explainability, and dashboard-based monitoring can provide a more usable fraud detection system than rule-based screening alone.
Keywords: UPI fraud detection, XGBoost, Random Forest, Explainable AI, SHAP, SMOTE, Streamlit dashboard, financial transaction security
Keywords: UPI fraud detection, XGBoost, Random Forest, Explainable AI, SHAP, SMOTE, Streamlit dashboard, financial transaction security
Abstract
Deep-SiamChange: A Multi-Scale Attention-Based Siamese Network for Robust Structural Change Detection in Urban Environments
D VIMAL KUMAR, A REVATHI, B YOGESHWARI
DOI: 10.17148/IJARCCE.2026.15420
Abstract: The automatic identification of structural changes in urban environments through bitemporal satellite imagery presents substantial challenges stemming from environmental noise, illumination variations, and the inherent complexity of distinguishing genuine construction alterations from transient phenomena. Traditional change detection methodologies frequently succumb to the "noise challenge," wherein variable sun angles, atmospheric interference, and seasonal vegetation fluctuations generate false positives that obscure authentic building modifications. This investigation introduces Deep-SiamChange, a novel architecture that integrates a Siamese encoder with multi-scale attention mechanisms and convolutional block attention modules to achieve time-invariant and noise-robust feature extraction. The proposed framework processes bitemporal imagery through twin neural pathways with shared weights, ensuring consistent feature extraction logic across temporal intervals. A Feature Pyramid Network captures structural details across multiple scales, enabling the detection of both minor residential extensions and substantial industrial developments. The integration of channel and spatial attention mechanisms filters environmental noise by emphasizing geometric structural patterns while suppressing illumination-related artifacts. Experimental evaluation on the LEVIR-CD benchmark dataset, comprising 637 high-resolution bitemporal image pairs, demonstrates that Deep-SiamChange achieves an F1-score improvement from 83.9% to 87.3% compared to baseline implementations. The architecture exhibits particular effectiveness in mitigating misregistration errors and maintaining detection accuracy under varying illumination conditions. These findings establish Deep-SiamChange as a practical solution for urban governance applications, including automated illegal construction monitoring, property tax assessment, and post-disaster structural assessment.
Keywords: Change Detection, Remote Sensing, Siamese Networks, Attention Mechanisms, Multi-Scale Feature Fusion, Urban Analytics
Keywords: Change Detection, Remote Sensing, Siamese Networks, Attention Mechanisms, Multi-Scale Feature Fusion, Urban Analytics
Abstract
EnergyLint: A Real-Time Static Code Analysis Engine for Energy-Efficient and Sustainable Software Development
Satyam Pandey, Nilesh Vishwakarma, Suraj Patel, Arya Mayekar, Prof. Vaishali Rane
DOI: 10.17148/IJARCCE.2026.15421
Abstract: Software engineering has long overlooked energy efficiency as a development-time concern, despite the growing carbon footprint of cloud-deployed code. Poorly structured algorithmic patterns - including nested iteration, unstructured recursion, and unnecessary sorting - introduce compounding CPU overhead that silently scales across distributed infrastructure. Existing approaches either intervene too late (post-deployment measurement) or create new energy burdens of their own (AI-assisted optimization). This paper introduces EnergyLint, a development-phase static analysis engine that identifies and remediates energy-costly code constructs directly within the IDE, relying exclusively on Python’s Abstract Syntax Tree (AST) module rather than runtime execution or language model inference. In controlled evaluation, EnergyLint achieved a 60.6% decrease in computed energy impact score, yielding an estimated power savings of 0.527 J (0.146 mWh) per execution and preventing 0.059 mg of CO₂ emissions per run - validated by direct Intel RAPL hardware measurement - demonstrating a viable, zero-dependency pathway toward sustainable software development.
Keywords: Green Software Engineering, Static Code Analysis, Abstract Syntax Tree, Energy Efficiency, Sustainable Computing, Code Optimization, Carbon-Aware Development, Shift-Left Sustainability
Keywords: Green Software Engineering, Static Code Analysis, Abstract Syntax Tree, Energy Efficiency, Sustainable Computing, Code Optimization, Carbon-Aware Development, Shift-Left Sustainability
Abstract
Sentiment Analysis Using RoBERTa-Based Hybrid Model
M. Srivalli, A. Sri Nagesh, Jaideep Gera, Sk. Moinuddin, P. Gnana Suji
DOI: 10.17148/IJARCCE.2026.15422
Abstract: The sentiment analysis of movie reviews is a well-known issue within the Natural Language Processing (NLP) field and finds the primary applications in opinion mining and movie recommendation systems. Over the last few years, a number of studies have investigated the concept of hybrid deep learning structures that integrate transformer encoders with recurrent and structured prediction stack to enhance sentiment classification. The transformer encoder in the given study is substituted with a Robustly Optimized BERT pretraining approach (RoBERTa) and LSTM with a Bidirectional Long Short-Term Memory (BiLSTM) network. The suggested method combines RoBERTa-BiLSTM and a Conditional Random Field (CRF) layer to maintain consistency with base architecture. The suggested framework is tested on the IMDb movie review data set with a structured train validation test splits. The evaluation of performance is done by applying standard classification measures. The model based on RoBERTa achieves an accuracy of 91.01, a precision of 91.43, a recall of 90.50 and an F1-score of 90.97, and the results were higher than the results reported on the Transformer- LSTM-CRF. These results imply that improved contextual representations supplied by the contemporary pre-trained transformers have a positive effect on document-level sentiment classification.
Keywords: Sentiment Analysis, Deep Learning (DL), RoBERTa, BiLSTM, CRF, IMDb Dataset.
Keywords: Sentiment Analysis, Deep Learning (DL), RoBERTa, BiLSTM, CRF, IMDb Dataset.
Abstract
An Efficient Image Classification System Using OpenCV and Deep Learning Models on Google Colab
Vishal Kumar, Kuldeep Chauhan, Varun Bansal, Sumika Jain
DOI: 10.17148/IJARCCE.2026.15423
Abstract: Image classification is a fundamental task in computer vision that enables machines to automatically categorize images into predefined classes with minimal human intervention. With the rapid advancement of deep learning techniques, particularly Convolutional Neural Networks (CNNs), the performance and accuracy of image classification systems have improved significantly [1], [2]. This paper presents the design and implementation of an efficient and scalable image classification system using Python, OpenCV, and Google Colab.
In the proposed approach, OpenCV is employed for image preprocessing tasks such as resizing, normalization, noise reduction, and color space conversion, which enhance the quality and consistency of input data. For classification, deep CNN architectures including ResNet and MobileNet are utilized [3], [5]. ResNet enables the training of deeper networks through residual learning, while MobileNet provides a lightweight architecture suitable for real-time and resource- constrained environments.
The model is trained and evaluated using GPU acceleration available in Google Colab, which significantly reduces computational time and improves training efficiency. The system is assessed using standard performance metrics such as accuracy, precision, recall, and F1-score to ensure comprehensive evaluation. Experimental results demonstrate that the proposed system achieves high classification accuracy while maintaining low computational complexity.
Furthermore, the integration of OpenCV preprocessing techniques with advanced deep learning models enhances feature extraction capability and overall system performance. The proposed framework is cost-effective, scalable, and easy to implement, making it suitable for a wide range of real-world applications, including healthcare diagnostics, security surveillance, and intelligent automation systems. This work highlights the effectiveness of combining traditional image processing techniques with modern deep learning approaches for robust image classification.
Keywords: Opencv, Google Colab, Resnet, Classification Accuracy, Healthcare Diagnostics, Convolutional Neural Networks (Cnns).
In the proposed approach, OpenCV is employed for image preprocessing tasks such as resizing, normalization, noise reduction, and color space conversion, which enhance the quality and consistency of input data. For classification, deep CNN architectures including ResNet and MobileNet are utilized [3], [5]. ResNet enables the training of deeper networks through residual learning, while MobileNet provides a lightweight architecture suitable for real-time and resource- constrained environments.
The model is trained and evaluated using GPU acceleration available in Google Colab, which significantly reduces computational time and improves training efficiency. The system is assessed using standard performance metrics such as accuracy, precision, recall, and F1-score to ensure comprehensive evaluation. Experimental results demonstrate that the proposed system achieves high classification accuracy while maintaining low computational complexity.
Furthermore, the integration of OpenCV preprocessing techniques with advanced deep learning models enhances feature extraction capability and overall system performance. The proposed framework is cost-effective, scalable, and easy to implement, making it suitable for a wide range of real-world applications, including healthcare diagnostics, security surveillance, and intelligent automation systems. This work highlights the effectiveness of combining traditional image processing techniques with modern deep learning approaches for robust image classification.
Keywords: Opencv, Google Colab, Resnet, Classification Accuracy, Healthcare Diagnostics, Convolutional Neural Networks (Cnns).
Abstract
AI-Powered Multimodal Emotion Recognition: A Zero-Shot Framework Using Large Multimodal Models for Real-Time Affective Computing
Vanitha A, Mohammed Roshan Akther K M
DOI: 10.17148/IJARCCE.2026.15424
Abstract: Contemporary emotion recognition systems predominantly depend on Convolutional Neural Network (CNN) classifiers pre-trained on constrained label sets, rendering them brittle in unconstrained real-world conditions. This paper presents an AI-powered multimodal emotion recognition framework that exploits the zero-shot reasoning capability of the Google Gemini 1.5 Flash Large Multimodal Model (LMM) to perform high-fidelity facial micro-expression analysis directly in a web browser. The system captures live video frames using the HTML5 MediaDevices API, performs client- side JPEG compression via the Canvas API, and transmits Base64-encoded image payloads to the Gemini inference endpoint, along with a carefully engineered multimodal prompt. Dynamic JSON schema enforcement ensures structured, type-safe responses that contain the detected emotion, a contextual explanation, bounding-box coordinates (normalized to a 0–1000 scale), and an affective quote. The front-end is implemented using React 19, Vite 6, TypeScript, Tailwind CSS, and Framer Motion, forming a production-grade, Decoupled, Multimodal Architecture. Experimental evaluation demonstrates a mean round-trip inference latency of 1.3 seconds on broadband connectivity, as well as the ability to detect fine-grained affective states well beyond the canonical "Big Six" categories, including complex states such as "Suppressed Anger" and "Cautiously Optimistic." The architecture adheres to a strict zero-storage privacy model compliant with GDPR and CCPA. Results confirm that cloud-based LMM prompting supersedes CNN edge models for nuanced emotion understanding while remaining accessible via commodity hardware.
Keywords: Affective Computing, Emotion Recognition, Large Multimodal Models, Gemini 1.5 Flash, Zero-Shot Learning, Facial Action Coding System, Micro-Expression Analysis, Prompt Engineering, Human-Computer Interaction, Real-Time Vision.
Keywords: Affective Computing, Emotion Recognition, Large Multimodal Models, Gemini 1.5 Flash, Zero-Shot Learning, Facial Action Coding System, Micro-Expression Analysis, Prompt Engineering, Human-Computer Interaction, Real-Time Vision.
Abstract
Cloud-Based IoT System for Real-Time Monitoring and Filtration of Industrial Wastewater for Multi-Purpose Reuse
SK. Kalesha Vali, M. Nivas, Ch. R. V. Durga Sai, N. Sanjeeva Reddy, G. Vijay Kumar
DOI: 10.17148/IJARCCE.2026.15425
Abstract: Industrial wastewater containing harmful chemicals, heavy metals, and suspended solids poses severe environmental risks if discharged without treatment. Traditional laboratory-based testing is time-consuming and lacks real-time capability. This paper presents a cloud-based IoT system for real-time monitoring and automated filtration of industrial wastewater using an Arduino UNO integrated with pH, turbidity, temperature (DS18B20), and TDS sensors. Sensor data is transmitted to the ThingSpeak platform via the ESP8266 Wi-Fi module for remote visualization. An L298N motor driver controls a water pump for automated filtration, and a buzzer activates when parameters exceed safe thresholds. Experimental results confirm effective detection before and after filtration with real-time cloud updates. The proposed system is cost-effective, scalable, and suitable for industrial wastewater management and environmental compliance.
Keywords: IoT, Arduino UNO, water quality monitoring, pH sensor, TDS sensor, ThingSpeak, ESP8266, industrial wastewater.
Keywords: IoT, Arduino UNO, water quality monitoring, pH sensor, TDS sensor, ThingSpeak, ESP8266, industrial wastewater.
Abstract
Automated Real-Time Bottle Defect Detection Using YOLOv8, BoT-SORT Tracking, and Audio Alerts
SHENBAGA GANESHAN S, Dr. C. KARPAGAVALLI, Dr E. MARIAPPAN, Dr M. KALIAPPAN
DOI: 10.17148/IJARCCE.2026.15426
Abstract: The quality assurance in the manufacturing process requires precise and accurate identification of defects in the product in real-time in order to avoid operational losses. Traditionally, the inspection process in a manufacturing environment relies on human observation or classical image processing techniques. These methods are more likely to be erroneous and time-consuming and cannot adapt to changing defect characteristics without major programming changes. In this context, the development of an automated bottle defect detection system using the YOLOv8 framework, BoT- SORT multi-object tracking algorithm, and audio alerts in real-time is proposed in this research. A dataset with five classes of bottle defects: cap, label, crumbled, no-cap, and not-crumbled, with a total of 1,250 samples, has been created and fine-tuned with the pre-trained YOLOv8n model with 80 epochs. A script test2.py has been written in order to run the model in real-time on both images and videos, drawing bounding boxes with their respective probabilities. Finally, the performance of the model has been evaluated with a score of 0.965 mAP@0.5, 0.953 macro F1, and 47 FPS on GPU.
Keywords: YOLOv8; Bottle Defect Detection; Machine Vision; BoT-SORT Tracking; Object Detection; Industrial Automation; Transfer Learning; Deep Learning; OpenCV; Ultralytics
Keywords: YOLOv8; Bottle Defect Detection; Machine Vision; BoT-SORT Tracking; Object Detection; Industrial Automation; Transfer Learning; Deep Learning; OpenCV; Ultralytics
Abstract
An Intelligent AI-Based Vehicle Breakdown Assistance System with Network-Aware Mobile Deployment
KARTHIKEYAN M, Dr C. KARPAGAVALLI, Dr E. MARIAPPAN, Dr M. KALIAPPAN
DOI: 10.17148/IJARCCE.2026.15427
Abstract: When a vehicle breaks down in the wilderness, finding a reliable connection to get help or assistance is often difficult or sporadic. For this reason, this study proposes an AI-based intelligent assistant, IntelliFix, to help users with broken-down vehicles through a Progressive Web Application (PWA)-based mobile system. The classification and semantic matching functions of the app are based on a transformer-based NLP. This enables accurate interpretation of the user's breakdown by gathering both data and integrating the user's account within the context of the app, regardless of whether or not the user was connected to the network. An additional aspect of the proposed Progressive Web Application (PWA) is its use of GPS and OpenStreetMap to geocode the location of broken-down vehicles and rank them in proximity to nearby repair shops. The application will enable users to report repairs via WhatsApp as well as make sequential voice calls to various mechanics. The classification process produced an accuracy of 91.05%, and the semantic matching process had an accuracy of 99.47%. In summary, the breakdown assistance system presents an increase in mobilization and efficiency of providing intelligent roadside assistance.
Keywords: Vehicle Breakdown Assistance, Artificial Intelligence, Natural Language Processing, Progressive Web Application, Network-Aware Systems, Semantic Matching, OpenStreetMap, Intelligent Transportation Systems
Keywords: Vehicle Breakdown Assistance, Artificial Intelligence, Natural Language Processing, Progressive Web Application, Network-Aware Systems, Semantic Matching, OpenStreetMap, Intelligent Transportation Systems
Abstract
Deep Learning Based Real-Time Sign Language Recognition
Prof. Minal Patil, Rhushabh Gaikwad, Rushikesh Ghogare, Shekhar Khandale, Roshan Avhad
DOI: 10.17148/IJARCCE.2026.15428
Abstract: Sign language is a vital medium of communication for individuals with hearing and speech impairments, but the lack of knowledge among non-signers creates barriers. This project proposes a real-time sign language recognition system that can detect alphabets (A-Z) and numerics (0-9) from webcam video. The system combines modern deep learning techniques with web technologies to provide an accurate, fast, and user-friendly solution. The frontend uses WebRTC to capture video streams directly in a browser, making the system platform-independent and usable with any standard laptop or external camera. The backend uses FastAPI with WebSockets to enable real-time communication between the browser and the deep learning model, ensuring low-latency predictions.
The recognition model integrates EfficientNet (transfer learning) for feature extraction, combined with a CNN+RNN to capture spatial and temporal patterns. An attention mechanism enhances performance by focusing on the most informative frames, while a GRU classifier predicts the final alphabet or number with high accuracy. Training and validation are carried out using benchmark datasets along with self-collected samples to ensure adaptability in real-world conditions. The system prototype displays recognized signs as text beneath the video feed, with emphasis on accuracy, robustness, and real-time performance for applications in education, healthcare, and accessibility services.
Keywords: Sign Language Recognition, Deep Learning, Computer Vision, Real-Time System, Gesture Detection, Neural Networks
The recognition model integrates EfficientNet (transfer learning) for feature extraction, combined with a CNN+RNN to capture spatial and temporal patterns. An attention mechanism enhances performance by focusing on the most informative frames, while a GRU classifier predicts the final alphabet or number with high accuracy. Training and validation are carried out using benchmark datasets along with self-collected samples to ensure adaptability in real-world conditions. The system prototype displays recognized signs as text beneath the video feed, with emphasis on accuracy, robustness, and real-time performance for applications in education, healthcare, and accessibility services.
Keywords: Sign Language Recognition, Deep Learning, Computer Vision, Real-Time System, Gesture Detection, Neural Networks
Abstract
Solar-IQ: An IoT Based Smart Solar Monitoring and Intelligence System
Rajshri Pote, Dhananjay Sharma, Vivek Sotra, Asmit Hade, Kapil Prajapati, Yash Chavhan
DOI: 10.17148/IJARCCE.2026.15429
Abstract: With the growing need for energy that is renewable, solar power has become increasingly important as a source of energy. The efficiency of solar panels can be hampered by elements such as dust, fluctuating temperatures, improper maintenance, and inadequate monitoring. In order to address the aforementioned problems, the present project will provide a system referred to as Solar-IQ: Smart Solar Intelligence System which is based on Internet-of-things (IoT).
The system is intended to monitor different factors concerning the solar panels' efficiency such as voltage, current, temperature, light intensity, and generation of power. The data will be captured, analyzed using the microcontrollers, and then sent to the monitoring platform from where it will be evaluated and visualization will be done. Alerts will be issued concerning abnormal performances and maintenance needs.
Apart from the alert function, the system will assist in energy production monitoring and system efficiency evaluation via a straightforward user interface. The monitoring will help in minimizing power loss as well as extending the lifespan of the solar panels.
The Solar-IQ system is an economical and efficient approach to managing solar energy smartly. The system can be applied in households, industrial applications, and commercial solar power projects. This research illustrates the ability to improve renewable energy systems through the integration of IoT and smart monitoring methods.
Keywords: Solar Power, Internet of Things (IoT), Smart Monitoring, Renewable Energy, Energy Efficiency, Predictive Maintenance.
The system is intended to monitor different factors concerning the solar panels' efficiency such as voltage, current, temperature, light intensity, and generation of power. The data will be captured, analyzed using the microcontrollers, and then sent to the monitoring platform from where it will be evaluated and visualization will be done. Alerts will be issued concerning abnormal performances and maintenance needs.
Apart from the alert function, the system will assist in energy production monitoring and system efficiency evaluation via a straightforward user interface. The monitoring will help in minimizing power loss as well as extending the lifespan of the solar panels.
The Solar-IQ system is an economical and efficient approach to managing solar energy smartly. The system can be applied in households, industrial applications, and commercial solar power projects. This research illustrates the ability to improve renewable energy systems through the integration of IoT and smart monitoring methods.
Keywords: Solar Power, Internet of Things (IoT), Smart Monitoring, Renewable Energy, Energy Efficiency, Predictive Maintenance.
Abstract
Railway Track Health Monitoring System Using IoT and ESP32: A Simulation-Based Approach on Wokwi Platform
Vladimir Illich Arunan V V, Ms. R. T. Charulatha
DOI: 10.17148/IJARCCE.2026.15430
Abstract: The railway transport system is of great importance due to its safety and capacity for performing efficiently. The purpose of this paper is to describe a get a way for monitoring the health of railway tracks by using an IoT device and an ESP32 microcontroller. This will be done using different types of sensors, including the MPU6050 Accelerometer, Ultrasonic, and DHT22 Temperature-Humidity Sensors. The proposed implementation is capable of measuring the vibration of the tracks and track buckling, detecting track defects, and monitoring the weather.
The data will be transmitted wirelessly to the ThinkSpeak cloud platform, which allows people to access the information in real-time. The Wokwi site was used for simulation of the proposed system. The results indicate that real-time monitoring of railway tracks is feasible, and indicate that the proposed system is capable of locating issues such as track buckling or cracks. Data will be sent out every 20 seconds (for a 50-meter stretch of track) by the proposed system for continuous tracking of the railway tracks and provides advanced warning to the rail operator about any potential issues with the railway tracks.
Keywords: Railway Track Monitoring, IoT, ESP32, MPU6050, ThingSpeak, Wokwi Simulation, Structural Health Monitoring, Wireless Sensor Networks
The data will be transmitted wirelessly to the ThinkSpeak cloud platform, which allows people to access the information in real-time. The Wokwi site was used for simulation of the proposed system. The results indicate that real-time monitoring of railway tracks is feasible, and indicate that the proposed system is capable of locating issues such as track buckling or cracks. Data will be sent out every 20 seconds (for a 50-meter stretch of track) by the proposed system for continuous tracking of the railway tracks and provides advanced warning to the rail operator about any potential issues with the railway tracks.
Keywords: Railway Track Monitoring, IoT, ESP32, MPU6050, ThingSpeak, Wokwi Simulation, Structural Health Monitoring, Wireless Sensor Networks
Abstract
COMPREHENSIVE REVIEW: CANCER TYPE DETECTION USING ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING
Prateek Sikarwar, Saurabh Singh, Aman Singh, Rohit Sharma
DOI: 10.17148/IJARCCE.2026.15431
Abstract: In recent years, the use of Artificial Intelligence in healthcare is increasing, mainly for detecting cancer. Early and correct diagnosis of cancer plays a crucial role for better treatment results and for lowering the number of deaths. This work describes a deep learning system that is designed to identify and categorize different cancers like brain tumor, skin cancer, lung cancer, and breast cancer, using medical imaging data.
The system uses Convolutional Neural Networks (CNNs), which helps in analyzing images and automatically extract important information from images. Different datasets are collected and organized into groups such as benign, malignant, and normal. Before training, images undergo preparation steps such as resizing, normalization, and data augmentation to improve results and reduce overfitting.
Different CNN models are trained for each cancer type with TensorFlow and Keras frameworks. The performance of these models is measured using metrics such as accuracy and loss. Experimental results show that some models achieved high accuracy (approximately 85–90%), while others demonstrated moderate performance due to challenges such as limited dataset size and class imbalance.
To enhance usability, a simple and interactive graphical user interface (GUI) is developed, allowing users to upload medical images and obtain real-time predictions along with confidence scores. Additionally, an invalid image detection mechanism is incorporated to prevent incorrect predictions for unrelated inputs, thereby improving system reliability.
Overall, this paper demonstrates the effectiveness of deep learning in cancer detection while also highlighting key challenges such as data limitations and model generalization. The proposed system can serve as a foundational framework for future research and can be further improved using advanced architectures, larger datasets, and real-time deployment strategies for practical healthcare applications.
Keywords: Cancer Detection, Medical Imaging, Artificial Intelligence, Machine Learning, Deep Learning, Convolutional Neural Networks (CNN), Healthcare Analytics, Tumor Classification, Clinical Decision Support Systems.
The system uses Convolutional Neural Networks (CNNs), which helps in analyzing images and automatically extract important information from images. Different datasets are collected and organized into groups such as benign, malignant, and normal. Before training, images undergo preparation steps such as resizing, normalization, and data augmentation to improve results and reduce overfitting.
Different CNN models are trained for each cancer type with TensorFlow and Keras frameworks. The performance of these models is measured using metrics such as accuracy and loss. Experimental results show that some models achieved high accuracy (approximately 85–90%), while others demonstrated moderate performance due to challenges such as limited dataset size and class imbalance.
To enhance usability, a simple and interactive graphical user interface (GUI) is developed, allowing users to upload medical images and obtain real-time predictions along with confidence scores. Additionally, an invalid image detection mechanism is incorporated to prevent incorrect predictions for unrelated inputs, thereby improving system reliability.
Overall, this paper demonstrates the effectiveness of deep learning in cancer detection while also highlighting key challenges such as data limitations and model generalization. The proposed system can serve as a foundational framework for future research and can be further improved using advanced architectures, larger datasets, and real-time deployment strategies for practical healthcare applications.
Keywords: Cancer Detection, Medical Imaging, Artificial Intelligence, Machine Learning, Deep Learning, Convolutional Neural Networks (CNN), Healthcare Analytics, Tumor Classification, Clinical Decision Support Systems.
Abstract
A DATA-DRIVEN MACHINE LEARNING FRAMEWORK FOR EMPLOYEE PRODUCTIVITY CLASSIFICATION IN WORK-FROM-HOME SETTINGS
Dr. Angelpreethi A, Gayathiri S, P Anitha
DOI: 10.17148/IJARCCE.2026.15432
Abstract: – The rapid transition to a work-from-home (WFH) culture has significantly transformed the traditional office environment, introducing new challenges in monitoring and evaluating employee productivity. Unlike conventional workplaces, remote work settings rely heavily on digital interactions, task-oriented workflows, and self-managed time, making productivity assessment more complex. In this study, a machine learning-based approach is employed to classify employee productivity levels in a WFH environment using a publicly available dataset. Work-related behavioral features are analyzed using supervised learning algorithms such as Decision Tree, K-Nearest Neighbors (KNN), and Naïve Bayes. The experimental results demonstrate that these algorithms can effectively classify employee productivity into predefined categories with satisfactory accuracy.
Keywords: Work From Home, Employee Productivity, Machine Learning, Classification, Remote Work.
Keywords: Work From Home, Employee Productivity, Machine Learning, Classification, Remote Work.
Abstract
Smart Gas Monitoring And Prediction System Using IOT And ML
M.Kumar, K. Mariya Priyadarsini
DOI: 10.17148/IJARCCE.2026.15433
Abstract: The dangerous effects of various forms of air pollution - as well as the presence of carbon monoxide (CO), ammonia (NH3), carbon dioxide (CO2), and other hazardous gases - pose a significant threat to both human health and the safety of our environment, especially in indoor spaces and in industrial settings. Because there are very few continuous monitoring systems, there is often a delay in detecting hazardous gaseous conditions, which increases the likelihood of accidents and negatively impacts health.This research project describes the design of a smart gas monitoring and prediction system that combines the Internet of Things (IoT) with Machine Learning (ML) to allow for the real-time monitoring and prediction of the concentrations of hazardous gases. This system is constructed using an ESP32 microcontroller, integrated with MQ-7, MQ-137, and SCD40 sensors to monitor CO, NH3, CO2, temperature, and humidity. All sensor data is captured and streamed to a ThingSpeak cloud service for storage, analysis, and visualization. To perform predictive analysis, a hybrid deep learning model consisting of Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) networks is constructed to analyze the time series data and forecast future levels of hazardous gases. This hybrid deep learning model is intended to enhance the accuracy of hazard detection and provide an earlier indication of dangerous conditions. A web-based dashboard has also been developed, which provides both the real-time and predicted values of the sensor measurements. Ultimately, the smart gas monitoring and prediction system will provide continuous monitoring of gaseous conditions, provide notifications of any hazardous conditions in a timely manner, and provide the opportunity to take action by implementing proactive safety measures; thereby improving the overall quality of environmental monitoring and increasing safety for the public.
Keywords: Internet of Things (IoT), Smart Gas Monitoring, Machine Learning (ML), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), ESP32, MQ-7 Sensor, MQ-137 Sensor, SCD40 Sensor, Carbon Monoxide (CO), Ammonia (NH₃), Carbon Dioxide (CO₂), ThingSpeak, Air Quality Monitoring, Predictive Analytics
Keywords: Internet of Things (IoT), Smart Gas Monitoring, Machine Learning (ML), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), ESP32, MQ-7 Sensor, MQ-137 Sensor, SCD40 Sensor, Carbon Monoxide (CO), Ammonia (NH₃), Carbon Dioxide (CO₂), ThingSpeak, Air Quality Monitoring, Predictive Analytics
Abstract
Digital Clinic Appointment and Scheduling System
Mrs. M. Khamar, S. Rajasri, S. Gayathri, V. N. V. Kamakshi, Y. Rohitha
DOI: 10.17148/IJARCCE.2026.15434
Abstract: The Digital Clinic Appointment and Scheduling System is a web application that is proposed to help in the management of hospital requirements by proving appointment with the doctor for the patients. This web application enables the patients to book, reschedule, and cancel and appointment through WhatsApp chatbot service that helps in doctor’s real-time schedule for booking. It is an automated solution with reduces the operational time thereby increasing the efficiency. Doctors can verify their schedules, appointments of all the working shifts and store the data along with the digital history of healthcare solutions. The patients can do the appointments through this web application or WhatsApp or emergency triage through online payment. Here the data will be stored on structured data base with all the medical histories there by reducing the manual efforts. It provides an interactive user interface by facilitating management of appointments through mobile devices. It is a completely capable alternative solution for managing healthcare solutions instead of conventional healthcare facilities. It involves encryption mechanisms to provide security to the patients data and flexible management of data for anywhere at any time.
Keywords: Digital Clinic System, Appointment Scheduling, WhatsApp Chatbot, Electronic Health Records (EHR)
Keywords: Digital Clinic System, Appointment Scheduling, WhatsApp Chatbot, Electronic Health Records (EHR)
Abstract
Smart Expense Tracking and Tax Estimation Web Application
Mrs. K. Tejaswi, V. Yoshitha, T. Mounika, Y. Himeswari, V. Kavya
DOI: 10.17148/IJARCCE.2026.15435
Abstract: Manual techniques for monitoring the personal finances, which comprise income, expenditure, and budget, are many. Because of such manual techniques, high operational time and efficiency errors are generated. In order to tackle all the above challenges of manual operations. A smart expenses tracking and tax estimation web app is proposed in this project. Users will be able to safely store their income and expenses data which also comprises tax percentage and several other features of total expenses amount and savings. The web app is to be developed using Angular JS as Frontend Technology. Spring Boot as backend and MySQL database. Other features that will be implemented in this web application are the receipts of expenses will be scanned through Optional Character Recognition (OCR) or otherwise, allows the users to input information regarding their expenses via voice command. The web application will enable giving a monthly, yearly budget, giving the spending pattern, giving alerts or notification when the budget limit has exceeded, the other features, apart from what has already mentioned, will be the estimation of taxes on expenses, based on the expenses carried out, through category filtering, via pictorial form.
Index Terms: Personal Finance Management, Expense Recording, OCR Technology, Voice Commands, Budget Warnings, estimation of Taxes.
Index Terms: Personal Finance Management, Expense Recording, OCR Technology, Voice Commands, Budget Warnings, estimation of Taxes.
Abstract
AI-Enhanced Student-Centric Learning Resources Catalogue and Management Portal with Integrated Chatbot Assistance
Mrs. J. Mounika, P. Nithin, S. Lokesh, P. Samuel Kowshik T. Siddharadha
DOI: 10.17148/IJARCCE.2026.15436
Abstract: Due to the increasing availability of digital resources, it has become difficult for students to efficiently access and manage academic resources. The AI-Enhanced Student-Centric Learning Resources Catalogue and Management Portal with Integrated Chatbot Assistance is a centralized platform for organizing and accessing academic resources like lecture notes, PDFs materials, video links, previous question papers, important topics, online tests and syllabus copy. Ai powered Chatbot provides instant academic support to the students to clarify their doubts and guide them to find their relevant resources and recommend important topics based on their syllabus and exam patterns. The AI-enhanced student resources portal is a web based application we use angular framework for frontend, spring boot for backend technology, MySQL for database and REST APIs for AI integration. The online test module enables students to check their knowledge on the particular topic with automated evaluation. The system enhances student engagement, reduces dependency on multiple platforms to collect their relevant resources in higher education institutions.
Index Terms: Artificial Intelligence, Student-Centric Learning, Resource Management, Chatbot, Personalized Learning.
Index Terms: Artificial Intelligence, Student-Centric Learning, Resource Management, Chatbot, Personalized Learning.
Abstract
Role-Based User Authentication, Authorization and Secure Content Delivery System
Mrs. V. Krishna Vijaya, V. Balaram, P. Srikanth, Y. Bharath, Sk. Roshan Zameer
DOI: 10.17148/IJARCCE.2026.15437
Abstract: When web systems change, keeping data private and controlling who uses what becomes more pressing. Most security tools handle just simple identification tasks but struggle with fine-level permissions or strong protection for digital content, especially across complex stacks. Because of these shortcomings, this work brings a new stateless security design aimed at large- scale content distribution and user authorization tied to roles. Technically, the solution leverages a Spring Boot backend to enforce API-level security via Spring Security, while the frontend utilizes Angular to manage modular routing and responsive design. Bridging these layers, JSON Web Tokens (JWT) are employed to facilitate stateless, encrypted communication, eliminating the reliance on server-side sessions. Apart from old-style fixed designs, this setup keeps the user interface separate from the backend API - making access rules match no matter where you interact. Because of that split, what comes out is something ready for real-world use: built to grow, using MySQL to lock down data securely, while following today’s top-tier security guidelines for businesses.
Abstract
Smart Action Oriented Student Feedback System
Mrs. B. Kalyani, V. Vardhan Babu, S. Syam Noel, P. Manideep Sai, R. Gowri Nikhil
DOI: 10.17148/IJARCCE.2026.15438
Abstract: The Full Stack Student Feedback Collection, Evaluation and Reporting Platform is a web-based Application designed to systematically collect, analyze, and present student feedback in an actionable format for academic improvement. The platform is developed using React-JS for the Client-side interface, Java Spring Boot for backend services, and a relational database for Persistent data storage, and Spring Boot reference Implementations by adapting core entities to Student and Feedback. Updating the system with various applications with the integration of an Action-Oriented Feedback Insight Engine, which is automatically, evaluates both quantitative ratings and quality textual feedback’s from the students. The engine uses sentiment classification, keyword-based model extraction, and trend analysis techniques to transform raw feedback into simple, actionable insights and understandable by the faculty members. Instead of presenting only numerical scores, this platform will generates clear recommendations highlighting strengths, areas for improvement where the faculty members are low, and recurring student feedback’s. The system ensures structured feedback of that particular faculty, efficient data processing and meaningful visualization of outputs from the students through faculty dashboards and analytical reports. By reducing manual interaction efforts and enabling data-driven from the platform to make a decision, this platform will supports the faculty to continuous improvement in teaching and providing a quality academic delivery. The proposed solution demonstrates a scal- able, real-world application of full stack development concepts and enhances traditional feedback systems by bridging the gap between feedback collection and effective action.
Index Terms: Java Spring Boot, ReactJS, MySQL, Action- Oriented AI-Bot
Index Terms: Java Spring Boot, ReactJS, MySQL, Action- Oriented AI-Bot
Abstract
User-Centric Bug Tracking Issue Logging and Resolution Workflow Management System
Mrs. K. Deepthi, Sk. Chinna Nagul Meera, P. Naga Babu, T. Gangadhara Rao, P. Venkata Hemanth Kumar
DOI: 10.17148/IJARCCE.2026.15439
Abstract: This paper is based on the web application that can be used to identify, manage, and resolve the important bugs in the software development. This system allows users to register the problems occurring in the software as bugs, also it tracks the status of the operation, and also it ensures that they are resolved in a particular time. As frontend this application uses the ReactJs Single Page Application, and also mainly it provides simple, fast and better experience for the user. On backend RESTful services are used by using Java Spring Boot, in this application there is a clear separation between the front end and the back end.
Index Terms: ReactJS SPA, Java Spring Boot, MySQL, REST- full Services, CRUD operations.
Index Terms: ReactJS SPA, Java Spring Boot, MySQL, REST- full Services, CRUD operations.
Abstract
QR -Based Event Management System with an Attendance Tracking System
Mrs. K. Tejaswi, T. Srikanth, P. Srinivas, SK. Md. Yaseen, and S. Arun Kumar
DOI: 10.17148/IJARCCE.2026.15440
Abstract: Managing events in educational institutions and organizations involves a vast number of participants in events, which produces large datasets to be maintained and stored. Organizing events involves class-to-class announcements, manual registrations, paper ticketing, gathering and maintaining attendance records, collecting feedback, certificates issuing and storing events details which are being conducted. Managing above activities requires more human effort and time also prone to human errors while storing this vast data. This project presents a web-based solution which helps to digitalize and efficiently manage all activities of events. We provide a centralized platform which consists Event dashboard to display events, make Digital Registrations for events, Digital QR tickets, Attendance Tracking Dashboard, Digital Feedback collection, Winner Announcements, Automated Certificate Generation for attendees and an efficient data storage to maintain event records useful for future purpose. For the development of applications, we are using HTML, CSS, ReactJS as front end, Spring Boot Frame-work and Core Java for Backend connectivity [11], and a MySQL relational database for data storage.
Abstract
Task-Oriented Productivity And To-Do Workflow Optimization Dashboard
Mrs. V. Krishna Vijaya, N. Mahesh, U. Abhinash, N. Bhairavanadh, N. Thrinadh Chowdary
DOI: 10.17148/IJARCCE.2026.15441
Abstract: The Task-Oriented Productivity and To-Do Workflow Optimization Dashboard is a full-stack web application which is created to help users organize and manage their daily tasks in an easy and efficient way. The users are allowed to add tasks, update details, set priorities, track progress, and mark tasks as completed in the system through a clean and responsive interface. The frontend is built using Angular as a Single Page Application, which ensures smooth navigation and quick interactions without reloading the page. The backend is developed using Java Spring Boot with a micro service approach, where REST APIs handle task operations, user actions, and application logic. All task- related data is stored securely in a MySQL database, providing reliable and structured data management. Docker is used to containerize the frontend, backend, and database, making the application easy to deploy, run, and maintain on different systems. This paper is simple to understand, beginner-friendly, and demonstrates modern full-stack development concepts while delivering a practical productivity management solution.
Index Terms: Angular, Single Page Application (SPA), Java Spring Boot, Micro service Architecture, RESTfull Web Services, MySQL, Docker, Full-Stack Web Application
Index Terms: Angular, Single Page Application (SPA), Java Spring Boot, Micro service Architecture, RESTfull Web Services, MySQL, Docker, Full-Stack Web Application
Abstract
Distributed Car Rental Reservation, Fleet Allocation and Customer Billing Management
Mrs. P. Saritha, N. Srinivasu, Sk. Mahabub Subhani, P. Sravan Kumar Reddy, V. Rambabu
DOI: 10.17148/IJARCCE.2026.15442
Abstract: This study explores the development and implementation of an online vehicle rental platform that streamlines reservations, user authentication, and vehicle management. The system utilizes Angular JS for the front-end, Node.js for back-end processing, and MySQL for database management. This research emphasizes the architecture, methodologies, and advantages of the system, ensuring optimal user experience, security, and scalability. A performance assessment confirms its efficiency in managing rental transactions. This paper presents the design and implementation of a distributed car rental management suite that integrates reservation management, fleet issue, and customer billing. The system leverages Spring Boot for backend RESTful services, Angular for client-side interaction, and MySQL for persistent storage. The proposed architecture ensures scalability, modularity, and reliability in managing vehicle availability, customer reservations, and billing workflows. Experimental evaluation demonstrates that the suite provides efficient transaction management, reduces booking conflicts, and enhances user experience through a responsive web interface. This paper presents the design and implementation of a distributed car rental management system for client purposes. This management system mainly focuses on key activities like reservation management, fleet allocation, customer billing, payment history. This is an advance development of online platform service. It explores the development and implementation of the online reservation and online vehicle rental platform. This system is mainly utilized with Angular for frontend and MySQL for database management and Spring Boot and Rest API’s for backend. This research can pay attention to words, the service mode of the application. The main advantage of the system is to ensure the user experience and security, scalability of applications. It also manages the transaction of the rental cars, it confirms the efficiency in managing the car rentals and transactions. The client interaction to book a car and make the transaction is very easy with a secure manner.
Index Terms: Car Rental System, Fleet Management, Distributed Architecture, Spring Boot, Angular, MySQL, REST APIs, Billing Automation.
Index Terms: Car Rental System, Fleet Management, Distributed Architecture, Spring Boot, Angular, MySQL, REST APIs, Billing Automation.
Abstract
Online Bookstore Product Catalog with AI-Powered Personalized Recommendations and Chat Support
Mrs. P. Aksha Dhanalakshmi, B. Lakshman, Sk. Sadath Ali, T. Sundar Dharahas, T. Karthik
DOI: 10.17148/IJARCCE.2026.15443
Abstract: The increasing demand for personalized digital experience in e-commerce platforms has increased the need for artificial intelligence to enhance user experience. This project presents the development of an “online bookstore product catalog with AI-powered personalized recommendations and chat sup- port”. The system allows users to browse books, place orders, and manage purchases through a user-friendly web interface. To enhance personalization, the application uses an AI-based recommendation system that suggests books to users based on their browsing behavior, purchase history, and product similarity. The backend of the system is developed using Spring Boot, Spring AI, and MySQL for secure and efficient data management. The frontend is implemented using Angular.
Key terms: Spring Boot, MySQL, Online Bookstore, e-commerce, AI Chatbot
Key terms: Spring Boot, MySQL, Online Bookstore, e-commerce, AI Chatbot
Abstract
Online Examination Question Bank, Quiz Attempt and Result Evaluation Portal
Dr. M. Purnachandra Rao, V. Srinivasa Rao, Thatha Hemanth, Surabi Sanjay, T. Chandu Vardhan
DOI: 10.17148/IJARCCE.2026.15444
Abstract: The Online Examination Question Bank, Quiz Attempt, and Result Evaluation portal is a web-based application development using a React.js frontend, Spring boot RESTful backend, and database to provide a complete digital examination management solution. The portal provides a complete digital examination management solution. This examination web application provides an efficient and user friendly portal for conducting examinations on the online environment. The examination can be created dynamically by selecting the questions from the question bank and it also defines the time limit and also it provides the evaluation results after completing the examination. Once the user submit their answer responses the backend would dynamically evaluates the examination based on the evaluation rules and it also displays the results so the user can view and review their results simultaneously. This online examination platform is the practical implementation of the full stack project and also it helps to the educational institutions and online learning platforms.
Index Terms: RESTful Architecture, Spring Boot REST API, Online line Quiz, Online Examination, React.js
Index Terms: RESTful Architecture, Spring Boot REST API, Online line Quiz, Online Examination, React.js
Abstract
AI-Based Glove for Hearing Impaired
Poorna Chandra Tejaswi, Dr Kavitha A S, Nikhith Gowda R, Pavan S, Rakshitha S P
DOI: 10.17148/IJARCCE.2026.15445
Abstract: The hearing-impaired community faces persistent communication barriers due to limited sign language fluency among the general population. Existing assistive solutions are predominantly camera-dependent, cost- prohibitive, or restricted to single-language support. This paper presents a low-cost, real-time assistive glove system that translates American Sign Language (ASL) gestures into speech with multilingual output capabilities. The proposed system employs five flex sensors (36 Hz sampling) integrated with an MPU6050 inertial measurement unit to capture temporal signing patterns. A quantized Bidi rectional Long Short-Term Memory (Bi-LSTM) model performs on-glove inference for low-latency recognition, achieving 92% accuracy across 36 gesture classes (A-Z, 1-10). Recognized English tokens are translated to Kannada, Hindi, French, and Spanish via cloud APIs, with text-to-speech output. Bidirectional interaction is enabled through speech capture, conversion to text, and display on an integrated OLED screen. The system emphasizes portability, affordability, and robustness for daily use, with experimental validation demonstrating conversational latency and day-long battery feasibility. This work contributes a comprehensive framework for inclusive communication in multilingual environments.
Keywords: Sign language recognition, assistive technology, Bi-LSTM, wearable sensors, multilingual translation,bidirectional communication, embedded machine learning, ESP32-S3.
Keywords: Sign language recognition, assistive technology, Bi-LSTM, wearable sensors, multilingual translation,bidirectional communication, embedded machine learning, ESP32-S3.
Abstract
Artificial Intelligence in Healthcare: A Comprehensive Survey on Disease Prediction
Buddareddy Sumeeth, Akula Sai Venkat, E Madhumitha, Aditya Santosh Naik, Dr. Muhibur Rahman T.R
DOI: 10.17148/IJARCCE.2026.15446
Abstract: In recent years, machine learning has found growing application in clinical settings, particularly for automating disease-level risk assessment tasks that were historically reliant on physician judgment alone. This paper surveys the current state of AI-driven healthcare systems, drawing on peer-reviewed publications indexed in IEEE Xplore, Springer, ScienceDirect, and PubMed. We reviewed work spanning core algorithmic approaches—including Random Forest, SVM, KNN, and various neural architectures—alongside applied systems for symptom triage, drug safety screening, and patient-facing health assistants. To bring structure to this diverse body of literature, we introduce a four-tier classification scheme organized around increasing system sophistication: from basic symptom-to-diagnosis mapping, through individualized care recommendations, into clinical support tooling, and finally toward fully integrated AI health platforms. Performance dimensions examined include classification accuracy, precision-recall balance, response latency, and scalability characteristics. A recurring pattern across reviewed studies is the absence of any single system that simultaneously covers real-time symptom intake, medication interaction checking, personalized guidance, and conversational AI interaction within one coherent architecture. We discuss the practical and theoretical implications of this gap and sketch directions for future work.
Keywords: Artificial Intelligence; Healthcare Systems; Disease Prediction; Machine Learning; Random Forest; Support Vector Machine; K-Nearest Neighbors; Neural Networks; Symptom Analysis; Drug Interaction; Medication Safety; Clinical Decision Support; Personalized Medicine; Healthcare Chatbot; Predictive Analytics.
Keywords: Artificial Intelligence; Healthcare Systems; Disease Prediction; Machine Learning; Random Forest; Support Vector Machine; K-Nearest Neighbors; Neural Networks; Symptom Analysis; Drug Interaction; Medication Safety; Clinical Decision Support; Personalized Medicine; Healthcare Chatbot; Predictive Analytics.
Abstract
HYBRID DISENTANGLED GRAPH CONTRASTIVE LEARNING FOR INTENT-AWARE RECOMMENDATION
VANITHA A, REVATHI A, DARSHINI P.G
DOI: 10.17148/IJARCCE.2026.15447
Abstract: Recommender systems play a vital role in modern digital platforms by delivering personalized content to users. However, capturing the dynamic and multifaceted nature of user intent remains a significant challenge. Traditional models rely on static user-item interactions and fail to disentangle the multiple latent factors that drive user behavior. This paper proposes a Hybrid Disentangled Graph Contrastive Learning (HDGCL) framework for intent-aware recommendation. The model constructs a user-item interaction graph and applies disentangled representation learning to separate latent factors such as long-term preferences, short-term behavioral patterns, and situational context. A hybrid contrastive learning mechanism is employed to enhance robustness and discriminability of learned embeddings. Contextual signals including time, location, and mood are incorporated to enable dynamic adaptation of user intent. Experimental results demonstrate that HDGCL consistently outperforms state-of-the-art recommendation baselines in Precision, Recall, and NDCG while improving diversity and interpretability.
Keywords: Hybrid Disentangled Graph Contrastive Learning, Intent-Aware Recommendation, Graph Neural Networks, Disentangled Representation Learning, Contrastive Learning, User Behavior Modeling, Context-Aware Recommendation
Keywords: Hybrid Disentangled Graph Contrastive Learning, Intent-Aware Recommendation, Graph Neural Networks, Disentangled Representation Learning, Contrastive Learning, User Behavior Modeling, Context-Aware Recommendation
Abstract
SMART ATTENDANCE BASED ON FACE RECOGNITION USING RASPBERRY PI AND GSM MODULE FOR HOME INTIMATION
N. Syam, T. Ravindra Sai, G. Chinna Jagapathi Swamy, K. M. Priyadarshini
DOI: 10.17148/IJARCCE.2026.15448
Abstract: The Smart Attendance System based on face recognition using the Raspberry Pi 4 and GSM module is an automated solution for attendance management. The system captures real-time facial images of students and identifies them using face detection and recognition algorithms. Attendance is recorded automatically with date and time, reducing manual effort and preventing proxy attendance. The system also integrates a SIM900A GSM Module to send SMS notifications to parents, improving communication and transparency. All data is stored in a database and managed through a web application, allowing administrators to monitor and control attendance efficiently. The system is cost-effective, reliable, and suitable for educational institutions.
Keywords: Face Recognition, Smart Attendance System, Raspberry Pi, GSM Module, Automated Attendance, Computer Vision, Deep Learning, YOLOv8, FaceNet, DeepFace, Embedded Systems, Real-Time Monitoring
Keywords: Face Recognition, Smart Attendance System, Raspberry Pi, GSM Module, Automated Attendance, Computer Vision, Deep Learning, YOLOv8, FaceNet, DeepFace, Embedded Systems, Real-Time Monitoring
Abstract
WIRELESS POWER TRANSFER FOR ELECTRIC VEHICLES (EV) CHARGING IN MATLAB
S. Muralikrishna, Mr.M.Rama Krishna
DOI: 10.17148/IJARCCE.2026.15449
Abstract: The technology for Electric vehicles (EVs) has become the one with the quickest rate of growth in the transportation sector. This is necessary to lessen environmental pollution, carbon dioxide emissions, and fossil fuel depletion. The primary problem with EVs, which use batteries as their primary energy source, is how they charge their batteries. The inductively coupled power transfer methodology, which is thought to be the most sophisticated and well- liked way of charging EVs, is the foundation of the model we provide in this study. The suggested system employs the series-series (SS) wireless power transmission topology, and a thorough discussion of the system's parameters is provided. The performance of the suggested system is briefly examined along with a theoretical study utilizing three simulated models on the MATLAB/SIMULINK platform. Additionally, the simulation findings of inductively coupled wireless power transmission are validated for various types of charging stations.
Keywords: Electric vehicles (EV), Series-Series topology (SS), Inductive coupling Simulink, Single-phase Wireless Power Transfer (WPT), Three-phase WPT.
Keywords: Electric vehicles (EV), Series-Series topology (SS), Inductive coupling Simulink, Single-phase Wireless Power Transfer (WPT), Three-phase WPT.
Abstract
Design and Development of Air Quality Monitoring and Alert Systtem using IoT
Dr. Ome Nerella, Tellakula Manideep
DOI: 10.17148/IJARCCE.2026.15450
Abstract: In the air, there are many dust particles and pollutant gases such as carbon dioxide and carbon monoxide that created an air pollution. The indoor and outdoor air pollution has brought the illness and harmful effect to human health. This creates a need for an IoT Alarm Air Quality Monitoring System to detect the dust particle, pollutant gases, temperature and humidity in the surrounding. The objective of this research work is to develop an indoor and outdoor air quality monitoring system for different air quality parameters (carbon dioxide and carbon monoxide), temperature, humidity, and dust concentration (air particle). Besides, the Node-RED dashboard and Android app are developed for real-time remotely applications in this system. The system performance is evaluated by testing the sensor used in the research work. In this research work, ESP-32, MQ7, MQ135, DHT22, and DSM501A are mainly used to develop the hardware. The MQTT is implemented as publish-subscribe network protocol to transfer the data as a message with the specific topic name. In the MQTT, the Node-RED dashboard, Android app and hardware are the MQTT client which are able to publish and subscribe the message. The Node RED dashboard acts as a live dashboard for monitoring and alarming purpose whilst the Android Studio is used to develop an Android app for the monitoring and alarm system in the smartphone. The Node-RED dashboard and Android app are able to display the data and notification message for different parameters on healthy or unhealthy level. The user can activate and deactivate the alarm system in the Node-RED dashboard or Android app as well
Keywords: Air Quality Monitoring, IoT, ESP32, MQ-7, MQ-135, MQTT, Node-RED, Real-Time Monitoring, Smart Alert System.
Keywords: Air Quality Monitoring, IoT, ESP32, MQ-7, MQ-135, MQTT, Node-RED, Real-Time Monitoring, Smart Alert System.
Abstract
An EfficientNet-B4 Based Medical Deepfake Detection in Healthcare Image Analysis
Mr. A. Azeem, K. Gowthami, B. Indhu, K. Pavani
DOI: 10.17148/IJARCCE.2026.15451
Abstract: Deepfake technology, powered by artificial intelligence and deep learning, can now create highly realistic fake images, audio, and videos. While this innovation has many uses, it also poses serious risks in healthcare, where medical images like X-rays and CT scans can be altered. Such manipulation may lead to wrong diagnoses, affecting patient safety and hospital operations. This study focuses on building a reliable deep learning approach to identify fake medical images. Two datasets—knee X-rays and lung CT scans—were prepared, preprocessed, and labeled as real or fake. The EfficientNet-B4 model was then applied to detect manipulations. Results show that the model performs very well, achieving high accuracy in both datasets, especially in knee X-ray images. It also maintains a good balance between speed and performance, making it suitable for real-time use. Overall, the study demonstrates that EfficientNet-B4 is an effective solution for detecting medical deepfakes quickly and accurately.
Index Terms: Medical deepfake image detection, deep learning, EfficientNet-B4, convolutional neural networks.
Index Terms: Medical deepfake image detection, deep learning, EfficientNet-B4, convolutional neural networks.
Abstract
A Sensory Glove With a Limited Number of Sensors for Recognition of the Finger Alphabet of Polish Sign Language and with voice
T. Yatesh, P. Victor paul, Y. Veeranjaneyulu, Mrs. T. Manogna
DOI: 10.17148/IJARCCE.2026.15452
Abstract: This research presents a novel sensory glove designed for recognizing the finger alphabet of Polish Sign Language (PSL) using a minimal number of sensors while maintaining high classification accuracy. Traditional data glove designs typically incorporate multiple sensors across all fingers, prioritizing recognition accuracy at the expense of ergonomics, affordability, and practical usability.
Hand gesture is one of the method used in sign language for non-verbal communication. It is most commonly used by deaf & dumb people who have hearing or speech problems to communicate among themselves or with normal people. Various sign language systems has been developed by many makers around the world but they are neither flexible nor cost-effective for the end users
Key words: analog-to-digital converter (ADC), inertial measurement unit (IMU), MPEG Audio Layer 3(mp3)
Hand gesture is one of the method used in sign language for non-verbal communication. It is most commonly used by deaf & dumb people who have hearing or speech problems to communicate among themselves or with normal people. Various sign language systems has been developed by many makers around the world but they are neither flexible nor cost-effective for the end users
Key words: analog-to-digital converter (ADC), inertial measurement unit (IMU), MPEG Audio Layer 3(mp3)
Abstract
EXPLAINABLE DEEP LEARNING-BASED INTRUSION DETECTION SYSTEM FOR WIRELESS SENSOR NETWORKS WITH REAL-TIME EDGE DEPLOYMENT
Chilka Sadhana, Mr K Appala Raju
DOI: 10.17148/IJARCCE.2026.15453
Abstract: Wireless Sensor Networks are being used more and more in areas like healthcare and smart cities. This makes them a big target for cyberattacks. There are some problems with the ways we currently detect these attacks. Firstly the systems have a time detecting the attacks that do not happen very often. Secondly it is hard to understand how these systems make their decisions. Lastly it is difficult to use these systems in time. This paper talks about a system for detecting cyberattacks in Wireless Sensor Networks that use Wi-Fi. The system uses a kind of computer program called a Convolutional Neural Network. This program was trained using a lot of data from a dataset called AWID-CLS-R. The people who made this system did three things to make it better. They made sure the system had an amount of examples of each type of traffic. They also used a tool called SHAP GradientExplainer to understand which features of the traffic are most important for detecting each type of attack. Lastly, they made a dashboard that can be used in time to detect attacks.The system was. It worked very well. It was able to identify attacks 99.3% of the time. The system is also very small so it can be used on devices that do not have a lot of power like the Raspberry Pi 4.The people who made this system found out that some features of the traffic are more important than others for detecting types of attacks. For example the feature called wlan.fc.order is very important for detecting Flooding and Impersonation attacks. The feature called wlan.seq is very important for detecting Injection attacks. Overall this system is an improvement over the systems that are currently being used. It can detect attacks well it can explain how it makes its decisions and it can be used in real time. Wireless Sensor Networks are used in areas, including healthcare, smart cities and industrial monitoring and this system can help keep them safe, from cyberattacks including Flooding, Impersonation and Injection.
Index Terms - Intrusion Detection System, Wireless Sensor Network, Convolutional Neural Network, Deep Learning, SHAP Explainability, Class Imbalance, Balanced Under sampling, AWID-CLS-R, IEEE 802.11, Wi-Fi Security, Real- Time Monitoring, Feature Selection, Network Traffic Classification, Edge Deployment.
Index Terms - Intrusion Detection System, Wireless Sensor Network, Convolutional Neural Network, Deep Learning, SHAP Explainability, Class Imbalance, Balanced Under sampling, AWID-CLS-R, IEEE 802.11, Wi-Fi Security, Real- Time Monitoring, Feature Selection, Network Traffic Classification, Edge Deployment.
Abstract
A COMPARATIVE STUDY OF GENRE-BASED SENTIMENT ANALYSIS
Dr. Angelpreethi A, Hepshiba Sherly P A, P Anitha
DOI: 10.17148/IJARCCE.2026.15454
Abstract: The emergence of multiple online platforms has generated a huge volume of user-generated movie reviews from various sources like IMDb, Kaggle repositories, and Twitter. Due to the unstructured nature and diversity in using language, manual analysis is hard to conduct. Sentiment analysis hence serves as an effective means of automatic opinion polarity detection from textual data. However, sentiment expression in movie reviews is pretty much influenced by genre and dataset characteristics. This paper hence represents a comparison study of genre-oriented sentiment analysis based on both the lexicon-based approach and the machine learning-based approach. In this context, sentiment classification tasks are performed on a number of movie genres such as Naïve Bayes, Support Vector Machine, and Random Forest. The performances of these approaches are compared across a number of datasets, and the result reports that machine learning-based methods usually tend to gain higher accuracy, particularly in informal social media data.
Keywords: Sentiment Analysis, Movie Reviews, Genre-Wise Analysis, Machine Learning, LexiconBased Approach, IMDb, Twitter.
Keywords: Sentiment Analysis, Movie Reviews, Genre-Wise Analysis, Machine Learning, LexiconBased Approach, IMDb, Twitter.
Abstract
Hybrid Adaptive Agentic Architecture for Autonomous Content Curation and Delivery: A Serverless Implementation for Real-Time Tech News Syndication
Yeshwathram D, Thaaiyumanavan G R, Vinod D
DOI: 10.17148/IJARCCE.2026.15455
Abstract: The exponential growth of unstructured online content has rendered manual curation and rule-based automation ineffective for consistent, high-quality social media posting. This paper presents a fully implemented Hybrid Adaptive Agentic Architecture that autonomously searches, summarizes, and posts the latest tech news to Twitter (X). The system comprises three specialized agents—News Search Agent, Summarization Agent, and Twitter Posting Agent—orchestrated by a single main Python script. A focused Hugging Face T5 summarization pipeline generates tweet-ready content with an Internal Quality Score (IQS), while a dedicated URLs tracking page ensures 100% deduplication. The entire pipeline is deployed using serverless computing on Google Cloud Run with two daily cron triggers (8:00 AM and 8:00 PM IST), enabling true 24/7 operation without any local hardware. A Reflection Agent provides continuous self-correction by tuning the IQS threshold based on performance telemetry stored in a dedicated persistent storage layer. A web front-end further allows custom genre prompting for on-demand posting. Experimental evaluation on live tech feeds demonstrates 98.7% posting success rate, zero duplication, and measurable quality improvement through adaptive tuning. The architecture overcomes the brittleness of traditional RPA while maintaining deterministic execution and transparent self-optimization.
Keywords: Agentic Architecture, Machine Learning Summarization, T5 Pipeline, Self-Correcting Systems, Serverless Computing, Content Curation, Persistent Storage Telemetry, Twitter API Integration
Keywords: Agentic Architecture, Machine Learning Summarization, T5 Pipeline, Self-Correcting Systems, Serverless Computing, Content Curation, Persistent Storage Telemetry, Twitter API Integration
Abstract
Advertising Across Eras Traditional vs Modern Media and Their Influence on Brand Equity
Manasvi Dubey, Gohel Roshani, Dr. Hiren Harsora
DOI: 10.17148/IJARCCE.2026.15456
Abstract: This study evaluates how traditional and modern advertising channels influence brand identity, trust, customer engagement, and long-term brand equity in an increasingly competitive and fragmented media landscape. It explores the evolution of advertising from conventional mass communication mediums such as television, print, radio, and outdoor media to contemporary digital platforms including social media, search engines, and programmatic advertising. The research highlights how traditional media contributes to credibility, emotional connection, and long-term brand recall, while modern media enables precision targeting, real-time engagement, performance tracking, and personalized consumer experiences.
The study adopts a descriptive research approach using secondary data, case study analysis, and comparative frameworks to understand the effectiveness of each medium. It also examines consumer behavior patterns, trust dynamics, and engagement levels across different demographic segments. Key industry examples and campaign analyses are used to demonstrate how leading brands strategically integrate both media types to maximize reach and impact.
Furthermore, the research develops an integrated advertising framework that emphasizes the importance of combining emotional storytelling with data-driven decision-making. The findings suggest that a hybrid advertising strategy not only enhances brand visibility but also strengthens customer relationships and improves return on investment. This study contributes to marketing literature by providing practical insights and strategic recommendations for businesses aiming to optimize their advertising efforts in the digital age while maintaining the trust and authenticity associated with traditional media.
The study adopts a descriptive research approach using secondary data, case study analysis, and comparative frameworks to understand the effectiveness of each medium. It also examines consumer behavior patterns, trust dynamics, and engagement levels across different demographic segments. Key industry examples and campaign analyses are used to demonstrate how leading brands strategically integrate both media types to maximize reach and impact.
Furthermore, the research develops an integrated advertising framework that emphasizes the importance of combining emotional storytelling with data-driven decision-making. The findings suggest that a hybrid advertising strategy not only enhances brand visibility but also strengthens customer relationships and improves return on investment. This study contributes to marketing literature by providing practical insights and strategic recommendations for businesses aiming to optimize their advertising efforts in the digital age while maintaining the trust and authenticity associated with traditional media.
Abstract
Voice Interview Analyzer
Mirza Daniyal Baig, Alfaiz Samani, Shehzan Shaikh, Anas Nakade, Alfiya Mulla, Zeeshan Khan
DOI: 10.17148/IJARCCE.2026.15457
Abstract: Our Voice Interview Analyzer, a project built using Python, is designed to make the initial hiring process smarter and more efficient. Think of it as an AI assistant for recruiters; it listens to a candidate’s spoken answers to standard interview questions, quickly converts their speech to text, and then dives deep into the content. Using natural language processing, it analyzes not just what the candidate says, but how they say it gauging their confidence through their tone, the richness of their vocabulary, there emotions when they talk, and how relevant their answers are. This system generates an objective, data-driven report that helps hiring managers save a lot of time while ensuring a fair and consistent screening process that can uncover promising candidates who could otherwise be overlooked.
Keywords: Speaker diarization, speaker recognition, speech separation, Emotion detection, speech clarity, Confidence Score, overall confidence, actionable feedback.
Keywords: Speaker diarization, speaker recognition, speech separation, Emotion detection, speech clarity, Confidence Score, overall confidence, actionable feedback.
Abstract
Intrusion Detection System for Smart Agriculture Using Deep Learning
Mr. M. Rama Krishna, M Tech, P. Keerthi Chandana, B. Varshita Lakshmi
DOI: 10.17148/IJARCCE.2026.15458
Abstract: Smart agriculture has rapidly adopted the use of Internet of Things (IoT) technologies that enhance the process of monitoring farm conditions and making decisions based on that. Nevertheless, IoT devices deployed in open and harsh environments are very susceptible to DDoS attacks and other forms of cyber intrusions. This project aims at developing a solution for intrusion detection in smart agriculture systems by applying a deep learning approach. Specifically, this work focuses on the development of an IDS based on a hybrid architecture involving BiGRU and LSTM architectures in order to perform analysis of sequence data and identify malicious operations. To achieve this goal, the intrusion detection system will be built as a web application implemented in Python Flask. The application allows uploading of datasets, training of the model and visualizing the results. TBPTT will be applied to optimize model training process. In this paper, we assess the performance of our model based on metrics such as accuracy, precision, recall and F1-score. Moreover, it should be noted that we calculate mapping between attack severity level and agricultural impact indicators such as water use, fertilizer efficiency and crop risk. Our experiments show promising results regarding accuracy of the model.
Keywords: Smart Agriculture, Internet of Things (IoT), Intrusion Detection System (IDS), Deep Learning, BiGRU, LSTM, TBPTT, Cybersecurity, DDoS Attack Detection, Network Security, Flask Web Application, Time-Series Analysis, Agricultural Impact Analysis1
Keywords: Smart Agriculture, Internet of Things (IoT), Intrusion Detection System (IDS), Deep Learning, BiGRU, LSTM, TBPTT, Cybersecurity, DDoS Attack Detection, Network Security, Flask Web Application, Time-Series Analysis, Agricultural Impact Analysis1
Abstract
Safe Park Sentry System using YOLOv8 Approach
Safe Park Sentry System using YOLOv, Approach, Abdul Sami Mansuri, Khan Ali Hamza, Yaheya Labbay, Koli Aradhya Umesh, Ali Naushadali Tangsal, Mr. Mohammed Sharique Maqsood Ahmed Shah
DOI: 10.17148/IJARCCE.2026.15459
Abstract: Traffic violations and road accidents are increasing rapidly due to the growth of vehicles and lack of efficient monitoring systems. Traditional traffic monitoring systems rely heavily on manual supervision, which is time- consuming, inefficient, and prone to human error.
This paper presents the Safe Park Sentry System, an intelligent traffic monitoring system using YOLOv8 and OpenCV. The system detects violations such as helmet absence, triple riding, overspeeding, and illegal parking in real time.
The system includes a graphical user interface (GUI), database storage, and emergency alert features. The proposed solution improves accuracy, reduces manual effort, and supports smart city infrastructure.
This paper presents the Safe Park Sentry System, an intelligent traffic monitoring system using YOLOv8 and OpenCV. The system detects violations such as helmet absence, triple riding, overspeeding, and illegal parking in real time.
The system includes a graphical user interface (GUI), database storage, and emergency alert features. The proposed solution improves accuracy, reduces manual effort, and supports smart city infrastructure.
Abstract
Development of Fluoride Detection and Alert System for Drinking Water
Khadar Basha Shaik, V. Akhilesh, S. Harshith, V. Rajeev Reddy, Dr. Ome Nerella
DOI: 10.17148/IJARCCE.2026.15460
Abstract: Excess fluoride concentration in drinking water is a major public health concern in many regions. While fluoride in small amounts is beneficial, prolonged exposure to high concentrations can cause dental and skeletal fluorosis. This is the development of a low-cost reagent-based fluoride detection and alert system using a colourmetric analysis approach. The system utilizes a fluoride-specific chemical reagent that produces a measurable color change proportional to fluoride concentration. A TCS34725 RGB color sensor captures the intensity variation under controlled white LED illumination. The ESP32 microcontroller processes the RGB values, estimates fluoride concentration through calibration mapping, and compares it with permissible limits. Real-time monitoring and alert notifications are implemented through IoT cloud integration.
Keywords: Fluoride Detection, Colourmetric Analysis, TCS34725, ESP32, IoT Monitoring, Water Quality, Reagent- Based Detection
Keywords: Fluoride Detection, Colourmetric Analysis, TCS34725, ESP32, IoT Monitoring, Water Quality, Reagent- Based Detection
Abstract
Early Disease Detection of Plants Using Deep Learning
THANGAMAHESWARAN V, Mr. P. VETRIVEL, Dr E. MARIAPPAN, Dr M. KALIAPPAN
DOI: 10.17148/IJARCCE.2026.15461
Abstract: Due to the estimated yearly loss of 20-40% of the total crop production, plant diseases are a serious threat to the world's food security. Early agronomic interventions are enabled by precise and timely identification of the diseases. This paper presents an end-to-end deep learning model for the early diagnosis and classification of plant leaf diseases. In the suggested model, the publicly available benchmark dataset, "PlantVillage," which contains more than 54,000 leaf images with 17 types of diseases affecting tomato, potato, and corn crops, is used for the development of the suggested model based on a specially designed three-block CNN model, namely "PlantCNN." In the suggested model, the accuracy of the results is improved by the development of a high accuracy classifier based on the refined EfficientNet-B0 model. In the suggested model, the "Random Horizontal Flipping" technique is used for the improvement of the robustness of the model. In the suggested model, the "Disease Severity Index" with Low, Medium, and High levels is calculated based on the HSV color thresholding method for the estimation of the visually infected part of the leaf. In the suggested model, the results, the classification accuracy, the segmented leaf image, and the fertilizer quantity for the diseases are presented with the index based on the type and severity of the diseases. In the suggested model, the results are presented based on the development of the end-to-end model, which is implemented as an interactive multilingual web application based on the Streamlit library. In the suggested model, the accuracy of the results exceeds 95% for the "PlantCNN" model, and the accuracy exceeds 97% for the "EfficientNet-B0" model, with the macro-averaged F1-score results exceeding 0.96 for all the test classes. The suggested model bridges the gap between real-world precision agriculture and high-quality deep learning research conducted in the lab.
Keywords: Leaf Segmentation; Precision Agriculture; Streamlit; Deep Learning; Convolutional Neural Network; Plant Village Dataset; Efficient Net; Transfer Learning; Plant Disease Diagnosis.
Keywords: Leaf Segmentation; Precision Agriculture; Streamlit; Deep Learning; Convolutional Neural Network; Plant Village Dataset; Efficient Net; Transfer Learning; Plant Disease Diagnosis.
Abstract
Smart Highway
Prof. Suchitra Patil, Akhil Dandoti, Piyush More, Soham Pathare, Shubham Nagotkar
DOI: 10.17148/IJARCCE.2026.15462
Abstract: Modern transportation systems face significant challenges such as traffic congestion, road accidents, inefficient resource utilization, and environmental concerns. To address these issues, the integration of intelligent technologies into highway infrastructure has become essential. Smart Highway systems provide a scalable and adaptive solution by combining Internet of Things (IoT), sensor networks, real-time monitoring, and data processing to enhance traffic management and road safety.
This research presents an intelligent system titled “Smart Highway: An Integrated Approach for Traffic, Lane, Irrigation, and Emergency Management.” The proposed system consists of multiple modules including Smart Traffic Management, Smart Lane Control, Smart Irrigation System, and Smart Emergency Response System. Various sensors and monitoring devices are deployed to collect real-time data related to vehicle movement, environmental conditions, and road incidents.
In addition, a web-based monitoring platform is developed that provides live camera views of highway conditions. This platform enables continuous observation of road activities and supports intelligent decision-making. A key feature of the system is ambulance detection, where the system identifies emergency vehicles using camera input and automatically facilitates lane clearance by adjusting traffic flow. This ensures faster movement of ambulances and reduces delays during critical situations.
The Smart Traffic and Lane Management modules dynamically regulate vehicle flow and optimize lane usage based on traffic density. The Smart Irrigation system uses environmental data such as soil moisture and weather conditions to automate water distribution efficiently. The Smart Emergency System further enhances safety by detecting accidents and triggering immediate alerts to emergency services and nearby vehicles.
The proposed system demonstrates an efficient, scalable, and intelligent highway management model that improves road safety, reduces congestion, and promotes sustainable resource usage. The results indicate that integrating smart technologies along with real-time monitoring platforms can significantly enhance transportation efficiency and support the development of future smart cities.
Keywords: Smart Highway, IoT, Traffic Management, Ambulance Detection, Lane Management, Smart Irrigation, Emergency System, Intelligent Transportation System
This research presents an intelligent system titled “Smart Highway: An Integrated Approach for Traffic, Lane, Irrigation, and Emergency Management.” The proposed system consists of multiple modules including Smart Traffic Management, Smart Lane Control, Smart Irrigation System, and Smart Emergency Response System. Various sensors and monitoring devices are deployed to collect real-time data related to vehicle movement, environmental conditions, and road incidents.
In addition, a web-based monitoring platform is developed that provides live camera views of highway conditions. This platform enables continuous observation of road activities and supports intelligent decision-making. A key feature of the system is ambulance detection, where the system identifies emergency vehicles using camera input and automatically facilitates lane clearance by adjusting traffic flow. This ensures faster movement of ambulances and reduces delays during critical situations.
The Smart Traffic and Lane Management modules dynamically regulate vehicle flow and optimize lane usage based on traffic density. The Smart Irrigation system uses environmental data such as soil moisture and weather conditions to automate water distribution efficiently. The Smart Emergency System further enhances safety by detecting accidents and triggering immediate alerts to emergency services and nearby vehicles.
The proposed system demonstrates an efficient, scalable, and intelligent highway management model that improves road safety, reduces congestion, and promotes sustainable resource usage. The results indicate that integrating smart technologies along with real-time monitoring platforms can significantly enhance transportation efficiency and support the development of future smart cities.
Keywords: Smart Highway, IoT, Traffic Management, Ambulance Detection, Lane Management, Smart Irrigation, Emergency System, Intelligent Transportation System
Abstract
IoT-Based Industrial Machine Monitoring, Energy Analysis, and Safety Automation System
V. Yasaswini, G. Vijay Kumar
DOI: 10.17148/IJARCCE.2026.15463
Abstract: The rapid growth of industrial automation and smart manufacturing has intensified the need for continuous, real-time monitoring of machine performance, energy consumption, and environmental safety parameters. Traditional industrial setups rely on manual inspection and SCADA-based systems that are reactive, labor-intensive, and incapable of providing instant automated responses. This paper presents an IoT-Based Industrial Machine Monitoring, Energy Analysis, and Safety Automation System aligned with Industry 4.0 concepts. The proposed system integrates a ZMPT004T voltage sensor, ACS712 current sensor, DHT11 temperature and humidity sensor, DS18B20 machine temperature sensor, and a flame detection sensor with an ESP32 microcontroller. Sensor data is continuously acquired, processed to compute voltage (V), current (A), power (W), and cumulative energy consumption (kWh), and displayed locally on a 20×4 I2C LCD. Simultaneously, all parameters are uploaded to the ThingSpeak cloud platform every 6 seconds for real-time graphical visualization and historical analysis. The Blynk mobile application provides a remote dashboard for machine monitoring and ON/OFF control via a relay module. Automatic safety mechanisms—triggered by over-temperature conditions (>40°C machine, >45°C industrial) or fire detection—immediately disconnect the machine and activate an audible buzzer alert. Experimental results confirm accurate multi-parameter sensing, reliable cloud transmission, effective LCD output in all operating modes including normal, fire alert, and temperature violation, and successful automated safety response. The system is cost-effective, easily deployable, and scalable for small, medium, and large industrial environments.
Keywords: IoT, Industry 4.0, Industrial Machine Monitoring, Energy Consumption Analysis, Safety Automation, ESP32, ThingSpeak, Blynk, DHT11, DS18B20, ACS712, ZMPT004T, Fire Detection, Predictive Maintenance, Smart Manufacturing.
Keywords: IoT, Industry 4.0, Industrial Machine Monitoring, Energy Consumption Analysis, Safety Automation, ESP32, ThingSpeak, Blynk, DHT11, DS18B20, ACS712, ZMPT004T, Fire Detection, Predictive Maintenance, Smart Manufacturing.
Abstract
FitBuddy: A Personalized Digital Fitness Trainer for Health Monitoring and Lifestyle Management
Sumit Rothe, Sanskruti Pednekar, Samiksha Ovhal, Saloni More, Sandeep Shinde, Kanda Kumaran Thevar
DOI: 10.17148/IJARCCE.2026.15464
Abstract: The rapid growth of sedentary lifestyles and associated health risks has led to increased interest in mobile health (mHealth) technologies that support fitness and well-being. This paper presents FitBuddy, a personalized digital fitness trainer designed to provide integrated services including workout planning, diet recommendations, body mass index (BMI) calculation, and calorie tracking. The system adopts a user-centered design approach to improve usability and engagement while leveraging personalization techniques to enhance adherence to fitness routines. Prior studies demonstrate that personalized mHealth applications significantly influence user motivation and long-term behavioral change by integrating goal setting, feedback, and adaptive recommendations. The proposed system combines these principles into a unified platform, offering accessibility and efficiency. Experimental observations indicate that integrated tracking and personalization improve consistency and user satisfaction. The study concludes that digital fitness systems such as FitBuddy can effectively bridge the gap between professional fitness guidance and everyday accessibility.
Keywords: mHealth, fitness application, personalization, calorie tracking, BMI, mobile health systems, health monitoring, digital fitness.
Keywords: mHealth, fitness application, personalization, calorie tracking, BMI, mobile health systems, health monitoring, digital fitness.
Abstract
Smart Agriculture System for Grape Leaf Disease Detection Using AI, Image Processing and Sensors
Prof. S. M. Bankar, Prof.Sweety Narula, Akshara R. Jadhav, Shravani G. Zurange, Bhagyashri V. Kalamkar
DOI: 10.17148/IJARCCE.2026.15465
Abstract: Modern agriculture faces growing challenges including crop diseases, inconsistent environmental conditions, and the absence of intelligent monitoring tools. Grape cultivation, a major commercial crop in Maharashtra and across India, is particularly vulnerable to diseases like Black Rot, Esca, and Leaf Blight, which can cause severe yield losses if not detected early. This paper proposes a Smart Agriculture System that combines IoT-based environmental sensing, Digital Twin technology, and Artificial Intelligence to address these challenges in a unified and practical platform. The system uses an ESP32 microcontroller interfaced with DHT11, soil moisture, and MQ135 sensors to collect real-time field data on temperature, humidity, soil conditions, and air quality. Sensor readings are wirelessly transmitted to a web- based dashboard that forms a live Digital Twin of the farm. In addition, a Convolutional Neural Network (CNN) trained on the PlantVillage grape leaf dataset allows farmers to upload leaf images and instantly receive disease classification results — identifying Healthy leaves, Black Rot, Esca, or Leaf Blight with approximately 94.7% accuracy. The overall system reduces dependence on manual field inspections, enables timely alerts, and supports informed decision-making for better crop management. Experimental results confirm the solution is cost-effective, scalable, and well-suited for real- world deployment in smart farming environments.
Keywords: Smart Agriculture, Grape Leaf Disease Detection, ESP32, Digital Twin, CNN, IoT, Image Processing, Precision Farming, PlantVillage Dataset
Keywords: Smart Agriculture, Grape Leaf Disease Detection, ESP32, Digital Twin, CNN, IoT, Image Processing, Precision Farming, PlantVillage Dataset
Abstract
Energy Consumption Prediction in Smart Homes using IoT and Machine Learning
Aditi Srivastava, Aditi, Anshika Tyagi, Chhavi Garg, Diya Jain, Usha Kumari, Satish Kumar Soni, Uruj Jaleel
DOI: 10.17148/IJARCCE.2026.15466
Abstract: This research paper presents an intelligent system for predicting energy consumption in smart homes using Internet of Things (IoT) and Machine Learning (ML) techniques. With the rapid growth of smart devices and automation, efficient energy management has become a critical challenge. The proposed system collects real-time data from IoT- enabled devices such as smart meters, temperature sensors, humidity sensors, and occupancy detectors. The collected data is pre-processed and analysed using machine learning algorithms including Random Forest, Support Vector Machine (SVM), and Linear Regression to forecast future energy consumption. The system not only predicts energy usage but also identifies consumption patterns to optimize energy efficiency and reduce wastage.
Experimental analysis shows that advanced ML models, particularly Random Forest, provide higher accuracy compared to traditional methods. The proposed approach can be effectively applied in smart homes, smart cities, and industrial environments to support sustainable and cost-efficient energy management.
Keywords: Internet of Things (IoT), Machine Learning, Energy Consumption Prediction, Smart Homes, Random Forest, Support Vector Machine (SVM), Linear Regression, Energy Efficiency, Smart Energy Management, Time-Series Forecasting.
Experimental analysis shows that advanced ML models, particularly Random Forest, provide higher accuracy compared to traditional methods. The proposed approach can be effectively applied in smart homes, smart cities, and industrial environments to support sustainable and cost-efficient energy management.
Keywords: Internet of Things (IoT), Machine Learning, Energy Consumption Prediction, Smart Homes, Random Forest, Support Vector Machine (SVM), Linear Regression, Energy Efficiency, Smart Energy Management, Time-Series Forecasting.
Abstract
SolarMap AI: An AI-Based System for Personalized Solar Planning
Rajeshree Chaudhari, Dnyaneshwari Sonawane, Sania Shinde, Prof. Satish Kuchiwale
DOI: 10.17148/IJARCCE.2026.15467
Abstract: With the rapid growth in energy demand and the urgent need for sustainable energy sources, solar power has emerged as one of the most promising renewable energy options in India. However, widespread rooftop solar adoption in Maharashtra remains below its potential, primarily due to the lack of simple, reliable, and personalized assessment tools accessible to non-technical users. This paper presents SolarMap AI, an intelligent web-based platform that leverages artificial intelligence to simplify and personalize the complete solar planning process. The platform integrates real hourly solar irradiance and temperature data from the NASA POWER API with a pre-trained Random Forest (RF) machine learning model to estimate annual energy output. It performs financial analysis by computing installation costs per Ministry of New and Renewable Energy (MNRE) benchmark rates, PM Surya Ghar Yojana government subsidies, return on investment (ROI), payback period, and 25-year degradation-adjusted net savings. A site suitability score (0-9) is calculated using a Multi-Criteria Decision Making (MCDM) Weighted Sum Model across four factors: Solar Resource (GHI), Payback Period, Bill Coverage, and Roof Condition. Experimental evaluation across four Maharashtra cities — Mumbai, Pune, Nagpur, and Kolhapur — validates the system. The RF model achieves an R² score of 0.9400 after hyperparameter tuning. Results demonstrate that SolarMap AI successfully provides end-to-end personalized solar planning, empowering ordinary users to make informed decisions about rooftop solar adoption.
Keywords: Solar energy, Random Forest, NASA POWER API, MCDM, Rooftop solar, Machine learning, PM Surya Ghar Yojana, Maharashtra, Renewable energy, Energy prediction, Financial analysis
Keywords: Solar energy, Random Forest, NASA POWER API, MCDM, Rooftop solar, Machine learning, PM Surya Ghar Yojana, Maharashtra, Renewable energy, Energy prediction, Financial analysis
Abstract
Civic Issue Reporting and Tracking
Sanskruti Maskar, Purva Patil, Rajshri Chavare, Vaishnavi Bharmal, Mr. Akshay Jeevankumar Shinde
DOI: 10.17148/IJARCCE.2026.15468
Abstract: The Civic Issue Reporting and Tracking System is a digital platform that provides an effective, transparent, and user-friendly method for reporting and managing civic issues. Its goal is to improve communication between citizens and local authorities. Users can use the system to report issues in real time, such as road damage, issues with waste management, outages in the water supply, and problems with public infrastructure. Complaints are properly documented, tracked, and resolved within a structured framework as a result of this. The system's design, development, and implementation are discussed in this paper, with a focus on how it can streamline the complaint management process and increase municipal authorities' accountability. Real-time status updates, complaint categorization, and automated notifications are just a few of the features offered by the platform, which makes use of cutting-edge web technologies. Additionally, it enables administrators to effectively monitor, assign, and resolve issues. Improvements in citizen engagement, transparency, and response time are evident from an empirical evaluation of the system. The system enhances service delivery and promotes participatory governance by bridging the gap between the public and governing bodies. The Civic Issue Reporting and Tracking System serves as a scalable and effective solution for smart city initiatives and digital governance.
Keywords: Issue Tracking, Digital Platform, Citizen Engagement, Transparency, Urban Management, Smart Cities, E- Government, Civic Issues, Complaint Management System.
Keywords: Issue Tracking, Digital Platform, Citizen Engagement, Transparency, Urban Management, Smart Cities, E- Government, Civic Issues, Complaint Management System.
Abstract
IoT and Machine Learning Based Smart Solar Energy Monitoring and Automatic Dual-Axis Solar Tracking System
G.Ravi, B.Lohitha
DOI: 10.17148/IJARCCE.2026.15469
Abstract: Solar energy is one of the most promising renewable energy sources, yet its efficient utilization remains challenging due to fixed panel orientations, lack of real-time performance monitoring, and absence of intelligent data analysis. This paper presents an IoT and Machine Learning (ML) based Smart Solar Energy Monitoring and Automatic Dual-Axis Solar Tracking System. The proposed system integrates an Arduino UNO microcontroller and an ESP32 module with voltage, current (ACS712), and Light Dependent Resistor (LDR) sensors to continuously measure solar panel output parameters and automatically adjust panel orientation for maximum sunlight absorption. Sensor data comprising voltage, current, power, and energy generation is transmitted to the ThingSpeak cloud platform via Wi-Fi for real-time remote visualization. A dual-axis tracking mechanism employing four LDR sensors and two servo motors ensures continuous alignment of the solar panel with the sun, enhancing energy capture by approximately 30–40% compared to fixed installations. A Linear Regression-based machine learning model developed in Python predicts daily and monthly energy production and estimates potential income, enabling proactive energy management. A Streamlit- based web dashboard provides an interactive interface for real-time and historical data analysis. Experimental results confirm accurate parameter detection, effective solar tracking, reliable cloud data transmission, and practical income prediction. The system is cost-effective, scalable, and suitable for residential rooftop installations, smart solar farms, and educational applications.
Keywords: IoT, Arduino UNO, ESP32, solar energy monitoring, dual-axis solar tracking, LDR sensor, ACS712, ThingSpeak, machine learning, linear regression, Streamlit dashboard.
Keywords: IoT, Arduino UNO, ESP32, solar energy monitoring, dual-axis solar tracking, LDR sensor, ACS712, ThingSpeak, machine learning, linear regression, Streamlit dashboard.
Abstract
IoT and Machine Learning Based Smart Health Monitoring System
Sidagam Geethika, P. Bose Babu
DOI: 10.17148/IJARCCE.2026.15470
Abstract: The rapid advancement of Internet of Things (IoT) and Machine Learning (ML) technologies has opened new avenues for developing intelligent and cost-effective healthcare monitoring systems. This paper presents the design and implementation of an IoT and Machine Learning based Smart Health Monitoring System capable of continuously monitoring multiple physiological and environmental parameters in real time. The proposed system integrates the MAX30102 sensor for heart rate and blood oxygen saturation (SpO₂) measurement, the DS18B20 sensor for body temperature monitoring, and the DHT11 sensor for ambient temperature and humidity sensing. An ESP32 microcontroller acts as the central processing unit, collecting sensor data and transmitting it to the ThingSpeak cloud platform via Wi-Fi for remote access and visualization. A 16×2 I2C LCD display provides immediate local readout of health parameters. A Random Forest machine learning algorithm, deployed through a Streamlit-based Python application, classifies the patient's health condition as Normal or Critical. Additionally, a GSM module (SIM900A) sends automated SMS alerts and voice calls to caregivers when critical conditions are detected. Experimental results demonstrate that the system achieves reliable real-time monitoring, accurate ML-based health classification, and effective emergency notification. The system is low cost, portable, and scalable, making it highly suitable for home healthcare, elderly monitoring, and remote medical applications.
Keywords: Internet of Things (IoT), Machine Learning, Health Monitoring System, ESP32, MAX30102, ThingSpeak, Random Forest, Remote Patient Monitoring, GSM Alerts.
Keywords: Internet of Things (IoT), Machine Learning, Health Monitoring System, ESP32, MAX30102, ThingSpeak, Random Forest, Remote Patient Monitoring, GSM Alerts.
Abstract
SOLAR ROOF TOP WIRELESS CHARGING STATION FOR ELECTRICAL VEHICLES
T.Anjaneyulu, G.Ravi
DOI: 10.17148/IJARCCE.2026.15471
Abstract: The transition to electric vehicles (EVs) is critical for reducing greenhouse gas emissions and mitigating climate change. To support this transition, the development of efficient and accessible charging infrastructure is paramount. In this project, we propose a solar rooftop charging station for EVs, integrating renewable energy sources to power vehicles sustainably. The system utilizes Arduino-based control mechanisms to manage power flow and ensure optimal charging performance. By harnessing solar energy, this solution not only reduces reliance on grid electricity but also promotes eco-friendly transportation. The integration of Arduino technology enhances flexibility and scalability, making it adaptable to various environments and user needs.
Keywords: Electric vehicles, Solar power, renewable energy, eco-friendly, grid electricity
Keywords: Electric vehicles, Solar power, renewable energy, eco-friendly, grid electricity
Abstract
BabyEase Cradle: Automated Infant Care System
Prof.Sweta Waghmare, Atharv Shirkar, Saeed Thakur
DOI: 10.17148/IJARCCE.2026.15472
Abstract: The BabyEase Cradle is an innovative, Internet of Things (IoT)-based automated cradle system designed to enhance the comfort, safety, and convenience of infants and parents alike. Utilizing a Raspberry Pi 4 as the central processing unit, the system integrates a suite of sensors to provide continuous environmental and physiological monitoring. Key parameters such as ambient temperature, humidity, and moisture are tracked via DHT11 and rain sensors, while a PIR sensor and an analog microphone (integrated with an ADS1115 ADC) detect infant movement and crying, respectively. The system employs intelligent automation to respond to the infant's needs: detection of a cry triggers a DC motor for gentle rocking and a buzzer to play soothing melodies, while a cooling fan activates automatically if temperatures exceed predefined safety thresholds. For remote supervision, a Flask-based web dashboard provides real- time sensor telemetry and a live video feed via a USB webcam. This interface also grants parents manual override capabilities for all actuators. Furthermore, the system ensures rapid response through instant mobile alerts via the Pushover notification service in the event of crying, motion, or adverse environmental changes. By reducing the demand for constant manual supervision and leveraging embedded systems for smart childcare, the BabyEase Cradle represents a significant advancement in intelligent, connected parenting solutions.
Keywords: Raspberry Pi, IoT (Internet of Things) Smart Childcare, Raspberry Pi 4, Embedded Systems Automated Cradle, Real-time Monitoring Sensor Fusion, Flask Web Dashboard, Remote Surveillance.
Keywords: Raspberry Pi, IoT (Internet of Things) Smart Childcare, Raspberry Pi 4, Embedded Systems Automated Cradle, Real-time Monitoring Sensor Fusion, Flask Web Dashboard, Remote Surveillance.
Abstract
Explainable AI Driven Multimodal Framework for Robust Spinal Muscular Atrophy Detection
Smit Mahesh Wani, Namdeo Baban Badhe, Neeta Patil
DOI: 10.17148/IJARCCE.2026.15473
Abstract: Spinal Muscular Atrophy is a serious hereditary disease which is a result of the mutations in the Survival Motor Neuron 1 gene (denotes the primary gene of the disease) which causes the gradual degeneration of motor neurons. Early diagnosis and prompt diagnosis is crucial to enhance patient outcomes and to facilitate timely medical care. Nevertheless, the conventional diagnostic techniques used to diagnose Spinal Muscular Atrophy are costly, lengthy and inaccessible and thus lead to late diagnosis. Moreover, there is a lack of effective integration of various clinical sources of data using current methods, and this reduces diagnostic accuracy and reliability.
To resolve these issues, this paper introduces a new multimodal deep learning system named Multimodal Attention-based Fusion Network, combining genetic information, medical images, electrophysiological measurements and clinical annotations. The model also includes attention mechanisms and the Explainable Artificial Intelligence based on SHAP to improve the interpretability and performance. The suggested framework was adopted and used by integrating various data modalities, such as survival motor neuron gene copy numbers, magnetic resonance images and ultrasound images, electrical impedance myography signals, and patient clinical history like age, functional score, and family history. The accuracy of 95.8 percent and the recall of 96.1 percent were demonstrated in an experimental evaluation, that was higher than traditional and single-modality methods, and the study was conducted in accordance with PRISMA guidelines, which guaranteed a systematic and validated research methodology.
Keywords: Spinal Muscular Atrophy, Multimodal AI, Explainable AI, MAF-Net, Deep Learning, Systematic Review
To resolve these issues, this paper introduces a new multimodal deep learning system named Multimodal Attention-based Fusion Network, combining genetic information, medical images, electrophysiological measurements and clinical annotations. The model also includes attention mechanisms and the Explainable Artificial Intelligence based on SHAP to improve the interpretability and performance. The suggested framework was adopted and used by integrating various data modalities, such as survival motor neuron gene copy numbers, magnetic resonance images and ultrasound images, electrical impedance myography signals, and patient clinical history like age, functional score, and family history. The accuracy of 95.8 percent and the recall of 96.1 percent were demonstrated in an experimental evaluation, that was higher than traditional and single-modality methods, and the study was conducted in accordance with PRISMA guidelines, which guaranteed a systematic and validated research methodology.
Keywords: Spinal Muscular Atrophy, Multimodal AI, Explainable AI, MAF-Net, Deep Learning, Systematic Review
Abstract
Trustify: An Intelligent Multimodal Framework for Fake Review Detection with Explainable AI
Shraddha Shirish Shah, Dr.Anil Vasoya, Pranjali Kasture
DOI: 10.17148/IJARCCE.2026.15474
Abstract: The proliferation of deceptive online reviews, now amplified by generative AI, has severely undermined consumer trust and market integrity, with fraudulent content estimated to exceed 30% on major platforms. Existing detection systems face critical limitations in explainability, cross-platform generalization, and real-time performance. This paper presents Trustify, a novel production-ready framework that integrates Adaptive Particle Swarm Optimization (APSO) with a hybrid Convolutional Neural Network (CNN) architecture to detect fake reviews. The system fuses multimodal features—textual (BERT, GPT-2), behavioral, temporal, and network-based signals—and incorporates SHAP and LIME for transparent, human-interpretable predictions. Evaluated on a large-scale dataset of 10,255 reviews from Amazon, Yelp, TripAdvisor, and IMDb, Trustify achieves 99.4% accuracy, 98.9% precision, 98.5% recall, and a 98.7% F1-score with sub-200ms latency. A PHP-based web interface enables real-time analysis and human-in-the-loop evaluation. By combining high accuracy, operational transparency, and practical deployability, Trustify bridges the gap between research and industry readiness, offering a meaningful advancement toward restoring trust in digital marketplaces.
Keywords: Fake review detection, convolutional neural networks, adaptive particle swarm optimization, explainable AI, multimodal feature engineering, e-commerce security, trust analytics
Keywords: Fake review detection, convolutional neural networks, adaptive particle swarm optimization, explainable AI, multimodal feature engineering, e-commerce security, trust analytics
Abstract
COMPACT SELF-DIPLEXING MIMO ANTENNA WITH IMPROVED ISOLATION CHARACTERISTICS
Mr. N. Bujji babu, K. Karthik, L. Sai, K. Sridhar, K. Kumar Raju
DOI: 10.17148/IJARCCE.2026.15475
Abstract: The design and analysis of a small multiple-input multiple-output (MIMO) antenna for wireless and Internet of Things applications operating in the 2.4 GHz to 2.7 GHz frequency range are presented in this research.The suggested antenna is created using a straightforward planar construction and examined utilizing several electromagnetic simulation tools, in contrast to traditional designs that depend on substrate integrated waveguide (SIW) structures and CST-based calculations Within the designated ISM band, the antenna is intended to provide dual-port operation with enhanced isolation and effective radiation characteristics. The design guarantees lower mutual coupling between antenna parts without requiring intricate decoupling structures by utilizing suitable slotting and efficient feed procedures. Throughout the operational frequency range, the suggested setup provides appropriate gain, steady radiation patterns, and good impedance matching. To verify the antenna behavior, performance metrics such S-parameters, return loss, isolation, and radiation characteristics are examined.The findings show that the antenna provides dependable performance appropriate for short-range wireless communication systems, such as Bluetooth, Wi-Fi, and Internet of Things devices.
Keywords: Compact Antenna , Self-Diplexing Antenna , MIMO Antenna , High Isolation , IoT Connectivity , Dual- Band Antenna, Microstrip Patch Antenna , Defected Ground Structure (DGS) , Mutual Coupling Reduction , HFSS Simulation ,Wireless Communication , Return Loss (S11) , Envelope Correlation Coefficient (ECC) , Gain Enhancement , 2.4 GHz / 5 GHz Bands.
Keywords: Compact Antenna , Self-Diplexing Antenna , MIMO Antenna , High Isolation , IoT Connectivity , Dual- Band Antenna, Microstrip Patch Antenna , Defected Ground Structure (DGS) , Mutual Coupling Reduction , HFSS Simulation ,Wireless Communication , Return Loss (S11) , Envelope Correlation Coefficient (ECC) , Gain Enhancement , 2.4 GHz / 5 GHz Bands.
Abstract
IoT-BASED COAL MINE WORKER SAFETY AND ENVIRONMENT MONITORING SYSTEM WITH CLOUD ANALYTICS
K. Chaitanya Varma, Mr. M. Rama Krishna
DOI: 10.17148/IJARCCE.2026.15476
Abstract: Coal mining remains one of the most hazardous occupations globally, with workers constantly exposed to toxic gas accumulation, extreme temperatures, and oxygen-deficient underground environments. Conventional safety systems rely on manual inspections and threshold-only alarms that trigger regardless of worker presence, leading to delayed responses and frequent false alarms. This work presents an IoT-Based Coal Mine Worker Safety and Environment Monitoring System that integrates real-time environmental sensing with RFID-based personnel tracking and cloud-connected analytics. The system uses the LPC1768 ARM Cortex-M3 microcontroller interfaced with an MQ135 air quality sensor, DHT11 temperature and humidity sensor, and an RC522 RFID module for worker identification. A worker-aware alerting logic activates the emergency buzzer only when hazardous thresholds are breached and at least one worker is confirmed inside the mine, eliminating false alarms during unmanned shifts. An ESP32 Wi-Fi module receives structured data from the LPC1768 over UART and publishes JSON payloads to the Zoho IoT cloud platform via MQTT over a secure TLS connection, enabling remote real-time monitoring and historical analytics. Experimental results confirm accurate multi-parameter sensing, reliable RFID-based worker tracking, and stable cloud data delivery across all tested conditions.
Keywords: IoT, LPC1768, ARM Cortex-M3, ESP32, MQ135, RFID, Zoho IoT, MQTT, Real-Time Monitoring, Coal Mine Safety.
Keywords: IoT, LPC1768, ARM Cortex-M3, ESP32, MQ135, RFID, Zoho IoT, MQTT, Real-Time Monitoring, Coal Mine Safety.
Abstract
AI ASSISTED AUTOMATIC TAMARIND PULP EXTRACTOR FOR DOMESTIC SCALE
Mrs S.Nila, S.Divya, R.Deenadalayan, R.Mothish
DOI: 10.17148/IJARCCE.2026.15477
Abstract: The AI-Assisted Automatic Tamarind Pulp Extractor for Domestic Scale is developed to simplify and modernize the traditional process of tamarind pulp extraction. Conventional methods are labor-intensive, time- consuming, and often lack hygiene, making them inefficient for regular household use. This project introduces a compact and intelligent machine that automates the extraction and separation of tamarind pulp and seeds while ensuring safety and consistency.The system uses a DC gear motor to drive a perforated drum that efficiently crushes tamarind and separates pulp from seeds. An Arduino-based control unit integrated with temperature, vibration, and current sensors monitors the system in real time. A camera module is also included to analyze pulp quality, while the AI system detects abnormalities and provides alerts or automatic shutdown for safe operation.
The machine is built using food-grade materials to maintain hygiene and is designed to be compact, energy-efficient, and cost-effective for domestic use. Experimental results show an average separation efficiency of about 97%, with a significant reduction in processing time and manual effort. This project highlights the potential of integrating artificial intelligence into household appliances for improved productivity, safety, and convenience.
Keywords: Artificial Intelligence (AI), Tamarind Pulp Extraction, Domestic Food Processing, Automation, Arduino-Based System, Sensor Monitoring, Smart Kitchen Appliance, Seed Separation, Food Hygiene
The machine is built using food-grade materials to maintain hygiene and is designed to be compact, energy-efficient, and cost-effective for domestic use. Experimental results show an average separation efficiency of about 97%, with a significant reduction in processing time and manual effort. This project highlights the potential of integrating artificial intelligence into household appliances for improved productivity, safety, and convenience.
Keywords: Artificial Intelligence (AI), Tamarind Pulp Extraction, Domestic Food Processing, Automation, Arduino-Based System, Sensor Monitoring, Smart Kitchen Appliance, Seed Separation, Food Hygiene
Abstract
Diabetic Retinopathy Classification Using Vision Transformer: A Strategy for Small Dataset Challenges
Mr K. Appala Raju, M. Pravallika, M. Bhumika, M.V.K.S. Harika
DOI: 10.17148/IJARCCE.2026.15478
Abstract: Early Detection of diabetic retinopathy, a complication of Vision loss in advance stages of diabetes, is essential to avoid permanent vision impairment. However, the automatic detection of diabetic retinopathy through medical image processing requires a large number of training data to build a model with good performance. This poses a challenge when working with small datasets as these models need large datasets to perform well on unseen data. Conventional Neural networks(CNNs), often fall short in capturing long-range dependencies, global pathological features across high resolution retinal images, leading to suboptimal performance in early-stage diagnosis. To address these limitations, this study proposes a Vision Transformer (ViT) model, designed to elevate DR severity classification (ranging from NO DR to Severe DR) by leveraging the self-attention mechanisms of transformer architectures. Vision Transformer(ViT) is a Deep learning architecture that generally requires large datasets for effective training. However, in this work, a smaller dataset is used because large medical datasets are difficult to access due to privacy and datasharing restrictions. The proposed approach utilizes a hierarchical structure where retinal fundus images from public (Kaggle) dataset APTOS 2109 dataset and a private (FGADR Website) dataset FGADR are divided into non-overlapping patches, embedded, and enriched with positional information. The proposal model achieves accuracy rates on threeclassification of 90% on FGADR and 86% on APTOS 2019 dataset. The model exhibits high performance, achieving a quadratic Weighted kappa (QWK) score of 0.93 on FGADR and 0.86 on APTOS. The proposed model demonstrates the good results to perform multi-class classification of DR using limited number of images.
Keywords: Diabetic Retinopathy (DR), Blindness/ Vision Loss Detection, Disease Severity Classification, Convolutional Neural Networks (CNNs), Vision Transformer (ViT).
Keywords: Diabetic Retinopathy (DR), Blindness/ Vision Loss Detection, Disease Severity Classification, Convolutional Neural Networks (CNNs), Vision Transformer (ViT).
Abstract
EcoReport Civic Complaint Management System
Lect. Satyajit Deshmukh, Vishwajeet Ukey, Avinash Chauhan, Ritick Raj Pandey, Md Islam, Abdul Slam, Dhruv Patel
DOI: 10.17148/IJARCCE.2026.15479
Abstract: Environmental complaint management systems in many regions suffer from inefficiencies arising from manual workflows, delayed responses, and limited transparency, which often lead to unresolved grievances and diminished civic engagement. This paper presents EcoReport, an intelligent, web-based environmental monitoring and complaint management system that aims to enhance responsiveness, accountability, and citizen participation. EcoReport leverages artificial intelligence for automated complaint classification, prioritization, and semantic analysis, enabling faster decision-making and more efficient resource allocation. A role-based access control model defines distinct operational layers for citizens, administrative authorities, and field officers, ensuring structured workflow execution across the system.
The system incorporates key capabilities such as GPS-based field verification supported by image evidence, offline data capture for use in low-connectivity areas, and automated notification mechanisms that provide real-time status updates to users. Implemented with a modern web stack, EcoReport combines a reactive frontend, a scalable backend architecture, and API-driven AI integration to deliver a cohesive and extensible platform. In contrast to traditional complaint systems, EcoReport minimizes manual intervention, strengthens traceability, and accelerates resolution cycles through workflow automation and intelligent data processing.
Keywords: Environmental Monitoring, Complaint Management System, Artificial Intelligence, Smart Governance, Role-Based Access Control, Workflow Automation, GPS Verification, Citizen Engagement, Web Application, Semantic Analysis.
The system incorporates key capabilities such as GPS-based field verification supported by image evidence, offline data capture for use in low-connectivity areas, and automated notification mechanisms that provide real-time status updates to users. Implemented with a modern web stack, EcoReport combines a reactive frontend, a scalable backend architecture, and API-driven AI integration to deliver a cohesive and extensible platform. In contrast to traditional complaint systems, EcoReport minimizes manual intervention, strengthens traceability, and accelerates resolution cycles through workflow automation and intelligent data processing.
Keywords: Environmental Monitoring, Complaint Management System, Artificial Intelligence, Smart Governance, Role-Based Access Control, Workflow Automation, GPS Verification, Citizen Engagement, Web Application, Semantic Analysis.
Abstract
BRAIN-COMPUTER INTERFACE SYSTEMS USING ARTIFICIAL INTELLIGENCE: A REVIEW OF EEG-BASED APPROACHES
Vidya K T, Rajeshwari N
DOI: 10.17148/IJARCCE.2026.15480
Abstract: Brain–Computer Interface (BCI) systems enable direct communication between the human brain and external devices by interpreting neural signals, primarily captured through Electroencephalography (EEG). With the increasing demand for assistive technologies, healthcare monitoring, and intelligent human–machine interaction, EEG-based BCI systems have gained significant attention in recent years. However, the inherent complexity, noise, and variability of EEG signals pose major challenges in achieving accurate and reliable signal classification.
Recent advancements in Artificial Intelligence (AI) and Machine Learning (ML), particularly deep learning techniques, have significantly improved the performance of EEG-based BCI systems. Models such as Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), hybrid CNN–LSTM architectures, and attention-based models have demonstrated strong capability in extracting spatial and temporal features from EEG signals.
This paper presents a comprehensive review of recent research developments (2020–2026) in AI-driven EEG-based BCI systems. It analyzes various models, techniques, and challenges, and highlights future research directions for improving system accuracy and real-time applicability.
Recent advancements in Artificial Intelligence (AI) and Machine Learning (ML), particularly deep learning techniques, have significantly improved the performance of EEG-based BCI systems. Models such as Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), hybrid CNN–LSTM architectures, and attention-based models have demonstrated strong capability in extracting spatial and temporal features from EEG signals.
This paper presents a comprehensive review of recent research developments (2020–2026) in AI-driven EEG-based BCI systems. It analyzes various models, techniques, and challenges, and highlights future research directions for improving system accuracy and real-time applicability.
Abstract
VISIBILITY ENHANCEMENT OF LESION REGIONS IN CHEST X-RAY IMAGES WITH IMAGE FIDELITY PRESERVATION
Mr N. Bujii Babu, Koilapu Preetham
DOI: 10.17148/IJARCCE.2026.15481
Abstract: Pneumonia diagnosis via chest X-ray (CXR) imaging remains challenging due to low-contrast lesion regions and inter-reader variability. This paper presents an integrated framework combining intelligent image enhancement with deep learning-based pneumonia classification to improve diagnostic accuracy and lesion visibility.The enhancement module employs a hybrid approach: Contrast Limited Adaptive Histogram Equalization (CLAHE) preprocessing followed by a custom Lesion-Aware Enhancement Network (LAEN) built on U-Net architecture. LAEN selectively amplifies pneumonia-indicative opacity and consolidation patterns while preserving structural integrity, optimized via a multi-component loss function combining perceptual loss, SSIM loss, and pixel-wise reconstruction loss.For classification, a modified DenseNet-121 architecture with attention mechanisms classifies enhanced X-rays into three categories: Normal, Bacterial Pneumonia, and Viral Pneumonia. Grad-CAM visualization generates interpretable attention maps to localize diseased regions for radiologist guidance.The system is evaluated on two publicly available datasets: the Kaggle Chest X-Ray dataset (~5,856 images) and the NIH CXR-14 dataset. The enhancement module achieves SSIM of 0.941 and PSNR of 38.5 dB, demonstrating excellent fidelity preservation. The classification model achieves 97.3% accuracy, 98.1% sensitivity, 95.8% specificity, and AUC-ROC of 0.991, substantially outperforming baseline methods including standard DenseNet-121 (94.2% accuracy) and standalone CNN approaches.Results demonstrate that combining intelligent lesion enhancement with attention-based deep learning creates a robust clinical decision-support tool for pneumonia detection, improving both diagnostic accuracy and interpretability for radiologists.
Keywords: Chest X-ray Imaging, Pneumonia Classification, Image Enhancement, Lesion Visibility, DenseNet-121, U-Net Architecture, Attention Mechanisms, Grad-CAM Visualization, SSIM, PSNR, Deep Learning, Automated Diagnosis.
Keywords: Chest X-ray Imaging, Pneumonia Classification, Image Enhancement, Lesion Visibility, DenseNet-121, U-Net Architecture, Attention Mechanisms, Grad-CAM Visualization, SSIM, PSNR, Deep Learning, Automated Diagnosis.
Abstract
Pipeline Maestro: Intelligent Orchestration of Build, Test and Deploy in CI/CD Systems
Prof. Diksha Bansod, Riya Mundrika Patel, Trupti Narhari Khairkar
DOI: 10.17148/IJARCCE.2026.15482
Abstract: CI/CD pipelines are key tools in today's software engineering projects. However, current pipelines are static, require manual effort, and lack intelligent decision-making. This paper proposes Pipeline Maestro, an intelligent orchestration tool integrating automation, failure prediction, and pipeline optimization. The proposed system reduces execution time, minimizes failures, and improves overall efficiency.
Keywords: CI/CD, DevOps, Pipeline Orchestration, Artificial Intelligence, Automation, Software Deployment
Keywords: CI/CD, DevOps, Pipeline Orchestration, Artificial Intelligence, Automation, Software Deployment
Abstract
SMART INTRUDER DETECTION
Shree Rakshana R, Akrithi RJ, Rashmi R, Ms. Charulatha. R.T
DOI: 10.17148/IJARCCE.2026.15483
Abstract: The motion detection–based security system, designed by the engineering team, is among the most essential security projects globally, created due to their focus on unauthorized intrusion during restricted time periods (Zhang et al., 2020). The project involves a layered response mechanism, and automated alerts reflect the connection between technology and safety, with elements such as monitoring, alerts, deterrence, and escalation (Patel & Singh, 2019). The designers used motion sensors for monitoring, detection, and vigilance, with repeating cycles from the surrounding environment and the specified night window (Kim & Lee, 2021), and an IoT-based communication module created smooth and slightly immediate notifications, where triggers outline motion and messaging changes to an audible alarm (Gubbi et al., 2013). The response phase contains deterrents such as high-intensity lights, audible alarms, and emergency contacts (Sharma et al., 2022), while the initial phase contains subtle alerts, with layers of automated responses contributing to the secure appearance of the household’s perimeter, operating from 23:00 to 04:20 (Alam et al., 2020). The system implies protection through automated alerts and deterrent actions, with delay for escalation to local security authorities or a police patrol (Borgia, 2014), comprising three stages with a primary contact and two additional emergency contacts, emphasizing residents’ safety and reduced risk of harm (Sivaraman et al., 2018).
Keywords: Internet of Things (IoT), Intruder Detection System, Arduino Uno, ESP8266, PIR Sensor, OV7670 Camera Module, Motion Detection, Smart Surveillance, Time-Based Automation, Alert Escalation Mechanism
Keywords: Internet of Things (IoT), Intruder Detection System, Arduino Uno, ESP8266, PIR Sensor, OV7670 Camera Module, Motion Detection, Smart Surveillance, Time-Based Automation, Alert Escalation Mechanism
Abstract
Personalized Skin Disease Consultant and Care Recommendation Using Lifestyle-Based Analysis
Mrs. S N Khandare, Pratik Birpan, Anup Sawai, Mayuri Pathade, Akanksha Gurav
DOI: 10.17148/IJARCCE.2026.15484
Abstract: In rural and semi-urban locations, many people have skin problems and diseases and it is difficult for them to see a dermatologist. Early recognition of these skin conditions means the patient can take preventative measures, and go to the doctor when required. This research discusses a new web-based system that will perform preliminary evaluations of skin conditions by using deep learning algorithms. A MobileNet based Convolutional Neural Network (CNN) will classify the various skin conditions, based on images provided by the user. MobileNet was selected for use in this web- based system due to it being a lightweight architecture so that only minimal processing power will be required on mobile devices and web browsers. The system also provides basic skin care suggestions and precautionary information so that users can see what their possible next steps for care may be. The model has been trained using and evaluated using a dataset containing multiple types of skin problems and diseases. Experimental evaluation of the model yielded a validation accuracy rate of 92.4%. The results demonstrate that lightweight deep learning models can be reliable and accessible for preliminary skin disease screening applications. Such a system can help provide early awareness of skin issues and encourage users to seek medical attention from a dermatologist if needed.
Keywords: Deep Learning, MobileNet, Convolutional Neural Network (CNN), Web-Based Healthcare System, Dermatological Screening, and Skin Disease Identification.
Keywords: Deep Learning, MobileNet, Convolutional Neural Network (CNN), Web-Based Healthcare System, Dermatological Screening, and Skin Disease Identification.
Abstract
AI-Based Research Paper Analyzer And Generator Using NLP And Machine Learning
Prof. Diksha Bansod, Shubham Gokul Jadhao, Himanshu Gulab Dhande
DOI: 10.17148/IJARCCE.2026.15485
Abstract: In recent years, the exponential growth of academic publications has created a demand for intelligent systems that can analyze and generate research content efficiently. This study presents an AI-Based Research Paper Analyzer and Generator that leverages Natural Language Processing (NLP) and Machine Learning techniques to automate tasks such as grammar correction, plagiarism detection, keyword extraction, and research paper generation.
The proposed system integrates TF-IDF-based keyword extraction, transformer-based semantic similarity models for plagiarism detection, and a Flask-based web application for real-time interactions with users. Additionally, the system includes a recommendation engine that retrieves relevant research papers using external APIs.
Experimental evaluation demonstrates that the system significantly reduces manual effort, improves writing quality, and provides structured outputs suitable for IEEE/IJARCCE format. The proposed framework offers a scalable and efficient solution for both students and researchers.
Keywords: NLP; Machine Learning; Research Paper Analysis; Text Summarization; TF-IDF; Plagiarism Detection
The proposed system integrates TF-IDF-based keyword extraction, transformer-based semantic similarity models for plagiarism detection, and a Flask-based web application for real-time interactions with users. Additionally, the system includes a recommendation engine that retrieves relevant research papers using external APIs.
Experimental evaluation demonstrates that the system significantly reduces manual effort, improves writing quality, and provides structured outputs suitable for IEEE/IJARCCE format. The proposed framework offers a scalable and efficient solution for both students and researchers.
Keywords: NLP; Machine Learning; Research Paper Analysis; Text Summarization; TF-IDF; Plagiarism Detection
Abstract
AUTOMATED RAILWAY SIGNALING SYSTEM WITH LOOPLINE FEEDBACK
Mr N. Bujii Babu, Janapala Balu
DOI: 10.17148/IJARCCE.2026.15486
Abstract: This project presents an Automated Railway Signaling System with Loopline Feedback designed to enhance railway safety and reduce human error at stations. The system is built around ESP32microcontroller board that coordinates multiple hardware components, including an LCD display, servo motor, IR sensors, limit switch, manual switches, buzzer, and a Wi-Fi module. Two manual switches are used by the station master to select the track and set the signal (red or green). Based on the switch status, a servo motor automatically changes the track alignment. A limit switch provides loopline feedback to confirm whether the track has been correctly changed, ensuring reliable operation. Two IR sensors are placed on the main line and loop line respectively to detect the presence of a train on either track. When the station master sets the signal to green, the system first verifies track alignment using the limit switch and checks for any existing train presence using the IR sensors. If the track is not properly set or a train is already present, the system prevents unsafe operation by activating a buzzer and displaying a warning on the LCD. Simultaneously, fault and status information is transmitted to a mobile application via the ThingSpeak cloud using internal Wi-Fi module. This integrated feedback and communication mechanism ensures real-time monitoring, improves operational safety, and demonstrates a low-cost, reliable approach to automated railway signaling and track management.
Keywords: Automated Railway Signaling System, Loop Line Feedback, Railway Safety, Track Switching, Signal Control, Train Detection, Real-Time Monitoring, Embedded System, IoT-based System
Keywords: Automated Railway Signaling System, Loop Line Feedback, Railway Safety, Track Switching, Signal Control, Train Detection, Real-Time Monitoring, Embedded System, IoT-based System
Abstract
Lung Cancer Detection Using Convolutional Neural Networks
P. Sampath Kumar, Dr. S. Mallikharjuna Rao*
DOI: 10.17148/IJARCCE.2026.15487
Abstract: Lung cancer is one of the leading causes of death worldwide, where early and accurate detection plays a critical role in improving patient survival rates. This paper presents an automated lung cancer detection system using Convolutional Neural Networks (CNN), a deep learning technique widely used for medical image analysis. The proposed system classifies lung CT scan images into three categories: Benign, Malignant, and Normal. The system incorporates image preprocessing techniques such as resizing, normalization, and noise reduction to enhance model performance. A CNN model is trained to automatically extract features and perform classification without manual intervention. Furthermore, the trained model is deployed using a Flask-based web application, enabling users to upload images and obtain real-time predictions along with confidence scores. Experimental results demonstrate that the proposed system achieves high classification accuracy and significantly reduces dependency on manual diagnosis. The system is efficient, cost-effective, and accessible, making it suitable for preliminary screening and decision support in healthcare environments. This work highlights the potential of deep learning in improving early-stage lung cancer detection and advancing medical diagnostic systems.
Keywords: Lung Cancer Detection, Convolutional Neural Network (CNN), Deep Learning, Medical Image Processing, Flask, Image Classification, Artificial Intelligence in Healthcare
Keywords: Lung Cancer Detection, Convolutional Neural Network (CNN), Deep Learning, Medical Image Processing, Flask, Image Classification, Artificial Intelligence in Healthcare
Abstract
An Optimized IoT-Based Coal Mine Safety Monitoring System with Edge-Driven Real-Time Hazard Detection
Bhavana L, Adithya Pratheep, Sarah Ninan, Vidya Sree, Shruthilayaa. M, Ms Charulatha R.T
DOI: 10.17148/IJARCCE.2026.15488
Abstract: Coal mining remains among the most dangerous occupations in the world, and the hazards underground are unlike those in virtually any other industry (Kowalski-Trakofler et al. 2011). Miners routinely work in the presence of toxic gases that build up silently in pockets of still air, in temperatures amplified by geothermal heat and continuous machinery operation, and in tunnels where radio signals fade and wired infrastructure is fragile at best. Underground conditions can deteriorate from normal to life-threatening in a matter of seconds, which leaves almost no margin for error in how the safety system responds. Despite improvements in wireless technology and cloud platforms, most current monitoring systems still rely on offloading decisions to remote servers and that round-trip delay is difficult to justify when the stakes are human lives.
This paper describes a monitoring system built around a different philosophy. Instead of centralizing the decision-making in the cloud, we move it onto the microcontroller itself, so sensor readings are evaluated and acted on before any network communication takes place. Gas levels and temperature are sampled continuously, and when a reading crosses a safety threshold, the device responds on its own: the buzzer goes off, the relevant LED lights up, and the event gets written to local storage. Cloud transmission is secondary — it happens after the local response, not as a precondition for it. The result is a system that reacts faster, draws less power, and keeps working normally even when connectivity is lost entirely.
Keywords: Internet of Things (IoT); Coal Mine Safety; Edge Computing; Gas Detection; Temperature Monitoring; Event-Driven Systems; Embedded Systems; Real-Time Monitoring; Industrial Safety.
This paper describes a monitoring system built around a different philosophy. Instead of centralizing the decision-making in the cloud, we move it onto the microcontroller itself, so sensor readings are evaluated and acted on before any network communication takes place. Gas levels and temperature are sampled continuously, and when a reading crosses a safety threshold, the device responds on its own: the buzzer goes off, the relevant LED lights up, and the event gets written to local storage. Cloud transmission is secondary — it happens after the local response, not as a precondition for it. The result is a system that reacts faster, draws less power, and keeps working normally even when connectivity is lost entirely.
Keywords: Internet of Things (IoT); Coal Mine Safety; Edge Computing; Gas Detection; Temperature Monitoring; Event-Driven Systems; Embedded Systems; Real-Time Monitoring; Industrial Safety.
Abstract
Security Issues in Electric Vehicle Charging Infrastructure
Ms. Disha S. Gopatwar, Mr. Kshitij P. Tajne, Mr. Piyush S. Borkhade, Prof. A. A. Gophane
DOI: 10.17148/IJARCCE.2026.15489
Abstract: This paper presents a comprehensive study on security issues in Electric Vehicle Charging Infrastructure (EVCI). With the rapid growth of electric vehicle adoption, charging stations have evolved into interconnected cyber- physical systems integrating communication networks, cloud platforms, and smart grid technologies [1], [9]. While these advancements enable smart charging, remote monitoring, and vehicle-to-grid (V2G) services, they also introduce significant cybersecurity risks [11], [13]. Vulnerabilities in communication protocols, firmware, authentication mechanisms, and backend management systems expose charging infrastructure to threats such as false data injection, denial-of-service attacks, malware intrusion, and unauthorized access [3], [10], [24]. These attacks can compromise user privacy, disrupt charging operations, and destabilize the power grid [2], [12]. This paper analyzes potential attack vectors and discusses security measures to enhance the resilience and reliability of EV charging systems.
Keywords: Electric Vehicle Charging Infrastructure, Cybersecurity, Smart Grid, Vehicle-to-Grid.
Keywords: Electric Vehicle Charging Infrastructure, Cybersecurity, Smart Grid, Vehicle-to-Grid.
Abstract
Special Minds Companion Assistant Using AI
Parth Khot, Pranav Akiwate, Rohit Babar, Pradnya Katkar, Shravani Jadhav, Shruti Hogade, Prof. K. S. Kadam
DOI: 10.17148/IJARCCE.2026.15490
Abstract: In today’s fast-paced world, individuals with special cognitive or developmental needs often face challenges in managing daily routines, maintaining focus, and achieving independent functioning. Traditional assistive tools provide limited personalization and lack continuous adaptive support. This paper proposes a Special Minds Companion Assistant, an AI-based intelligent system designed to provide personalized assistance, emotional support, and adaptive learning for individuals with conditions such as Autism Spectrum Disorder (ASD), ADHD, and learning disabilities. The system integrates Artificial Intelligence techniques such as Natural Language Processing (NLP), Machine Learning (ML), and speech recognition to deliver interactive and user-friendly assistance. It supports task scheduling, reminders, emotional wellness guidance, and caregiver monitoring. The system aims to enhance independence, improve routine management, and provide structured guidance, thereby contributing to better quality of life and learning outcomes.
Keywords: Artificial Intelligence, NLP, Special Needs, Assistive System, Machine Learning
Keywords: Artificial Intelligence, NLP, Special Needs, Assistive System, Machine Learning
Abstract
AI-Enhanced SMS Spam Detection Using Hybrid NLP and Machine Learning Techniques
Manish Singh Gahlot, Divyanshu Sharma, Bharat Saini, Dhananjay Kumar, Gagan Sharma, Ajit Singh, Satish Kumar Soni, Uruj Jaleel
DOI: 10.17148/IJARCCE.2026.15491
Abstract: Short Message Service (SMS) spam has emerged as a significant cybersecurity threat due to the rapid growth of mobile communication systems. With billions of SMS messages exchanged daily, malicious actors exploit this platform to distribute phishing links, fraudulent advertisements, fake financial alerts, and malware. Traditional rule-based spam filtering techniques, which rely on predefined keywords and patterns, have become ineffective against modern spam strategies that continuously evolve [1].
Recent advancements in Machine Learning (ML), Deep Learning (DL), and Natural Language Processing (NLP) have significantly improved the capability to detect spam messages with higher accuracy. Transformer-based models such as BERT further enhance semantic understanding of short text messages [10].
This research proposes a Hybrid Adaptive SMS Spam Detection Model that integrates TF-IDF, word embeddings, and transformer-based contextual representations. Additionally, it incorporates adaptive learning mechanisms to handle concept drift and evolving spam patterns. The proposed system not only improves classification accuracy but also reduces false positives, ensuring better user experience and system reliability.
Keywords: SMS Spam Detection, Machine Learning, NLP, BERT, Cybersecurity.
Recent advancements in Machine Learning (ML), Deep Learning (DL), and Natural Language Processing (NLP) have significantly improved the capability to detect spam messages with higher accuracy. Transformer-based models such as BERT further enhance semantic understanding of short text messages [10].
This research proposes a Hybrid Adaptive SMS Spam Detection Model that integrates TF-IDF, word embeddings, and transformer-based contextual representations. Additionally, it incorporates adaptive learning mechanisms to handle concept drift and evolving spam patterns. The proposed system not only improves classification accuracy but also reduces false positives, ensuring better user experience and system reliability.
Keywords: SMS Spam Detection, Machine Learning, NLP, BERT, Cybersecurity.
Abstract
AUTOMATED VERTICAL GARDENING SYSTEM USING IoT-ENABLED SMART IRRIGATION FOR SUSTAINABLE URBAN GREEN SPACES
Dr. Bhagyashree Dharaskar, Chandan Lokhande, Mohit Tembhurne, Kunal Patle, Mohit Goupale, Gaurav Hiwase
DOI: 10.17148/IJARCCE.2026.15492
Abstract: Cities are growing really fast these days, and because of that, we’re losing a lot of green spaces. This is causing problems like increased heat, bad air quality, and even affecting the natural balance around us. To deal with limited space, many people have started using vertical gardens. They are a good solution, but the main problem is maintaining them regularly. Not everyone has the time or proper knowledge to take care of plants every day.
So, in this project, I tried to make things easier by building an IoT-based vertical gardening system using NodeMCU. The idea is simple — instead of manually watering plants, the system does it automatically based on the plant’s actual needs. Sensors are used to check things like soil moisture and other conditions, and watering happens only when required. I tested this system for about 30 days, and the results were actually pretty good. Water usage was reduced by around 44%, and the effort needed for maintenance dropped a lot, almost by 90%. The system also worked reliably most of the time, with around 98.7% accuracy, which means it can be used in real-life situations like homes, offices, or even small public setups. I also created an Android app to make it easier to monitor everything. Through the app, users can see live data like soil moisture, temperature, humidity, and pH levels. This helps in understanding how the plants are doing without checking them physically again and again. There are two modes in the app — automatic and manual. So if someone wants full control, they can switch to manual and manage things themselves. Users can also set their own limits, like when watering should start or stop. If something goes wrong, like low moisture or any system issue, the app sends a notification immediately.
Overall, this project is a small step towards making gardening easier in cities. It connects simple hardware with a mobile app to reduce effort and save resources. It’s not too complicated and can be improved or expanded further, which makes it useful for creating greener spaces in urban areas.
Keywords: Smart Vertical Farming, Internet of Things (IoT), Automated Irrigation Systems, NodeMCU Controller, Environmental Monitoring, Urban Sustainability, Smart Agriculture Solutions.
So, in this project, I tried to make things easier by building an IoT-based vertical gardening system using NodeMCU. The idea is simple — instead of manually watering plants, the system does it automatically based on the plant’s actual needs. Sensors are used to check things like soil moisture and other conditions, and watering happens only when required. I tested this system for about 30 days, and the results were actually pretty good. Water usage was reduced by around 44%, and the effort needed for maintenance dropped a lot, almost by 90%. The system also worked reliably most of the time, with around 98.7% accuracy, which means it can be used in real-life situations like homes, offices, or even small public setups. I also created an Android app to make it easier to monitor everything. Through the app, users can see live data like soil moisture, temperature, humidity, and pH levels. This helps in understanding how the plants are doing without checking them physically again and again. There are two modes in the app — automatic and manual. So if someone wants full control, they can switch to manual and manage things themselves. Users can also set their own limits, like when watering should start or stop. If something goes wrong, like low moisture or any system issue, the app sends a notification immediately.
Overall, this project is a small step towards making gardening easier in cities. It connects simple hardware with a mobile app to reduce effort and save resources. It’s not too complicated and can be improved or expanded further, which makes it useful for creating greener spaces in urban areas.
Keywords: Smart Vertical Farming, Internet of Things (IoT), Automated Irrigation Systems, NodeMCU Controller, Environmental Monitoring, Urban Sustainability, Smart Agriculture Solutions.
Abstract
SmartBin: An IoT-Based Intelligent Waste Classification and Automated Segregation System Using Machine Learning
Shri Nivetha R, Kavya S Prasad, Jeyaarthi V, Hasanthika Sri S, Deepashri P M, Ms Charulatha R T
DOI: 10.17148/IJARCCE.2026.15493
Abstract: The rapid growth of urbanization has led to a significant increase in waste generation creating a need for efficient and intelligent waste management systems. Traditional methods of waste segregation depend on human labour, which is often inconsistent, labour-intensive, and error-prone. SmartBin, an Internet of Things (IoT) based automated waste classification and segregation system is proposed in this paper. The system employs an ESP32-CAM module to capture images of waste materials, which are then processed and classified by a machine learning model developed using TensorFlow running on a laptop server. Based on the classification outcome, a servo motor is rotated to direct the waste into either organic or recyclable compartments. Additionally, a real-time dashboard is implemented to visualise waste data, display captured images with their predicted labels and provide an option to export the data in CSV format. The proposed system provides a cost effective and scalable approach to automated waste segregation and thus contributing to the development of sustainable smart city.
Keywords: Internet of Things; Waste Segregation; Machine Learning; Computer Vision; ESP32-CAM; Smart Cities; Automation
Keywords: Internet of Things; Waste Segregation; Machine Learning; Computer Vision; ESP32-CAM; Smart Cities; Automation
Abstract
Explainable Artificial Intelligence-Based Network Intrusion Detection System Using SHAP, LIME and Counterfactual Analysis
Manam Siva Sai, Dr. Chandra Sekhar Koppireddy
DOI: 10.17148/IJARCCE.2026.15494
Abstract: The rapid evolution of cyber threats has exposed the limitations of traditional signature-based intrusion detection systems. While machine learning offers strong detection capability, its opacity undermines analyst trust. This research proposes an Explainable AI-based Network Intrusion Detection System combining XGBoost classification with a comprehensive SHAP explanation framework.The system processes traffic from three benchmark datasets — UNSW- NB15, CIC-IDS-2017, and NSL-KDD — through automated preprocessing covering encoding, normalisation, stratified partitioning, and class imbalance handling. The XGBoost classifier achieves F1-Scores of 0.9793, 0.9914, and 0.9853 respectively. SHAP TreeExplainer generates six visualisation types spanning global and local explanations, further complemented by LIME surrogate modelling and counterfactual generation — forming three mutually validating interpretability channels.Findings consistently identify volumetric flow statistics, behavioural connection-count features, and protocol-state indicators as dominant discriminating factors, aligning with established network security knowledge and reinforcing both the model's reliability and real-world applicability.
Keywords: Explainable AI, Network Intrusion Detection, SHAP, XGBoost, LIME, Cyber Security
Keywords: Explainable AI, Network Intrusion Detection, SHAP, XGBoost, LIME, Cyber Security
Abstract
Artificial Intelligence Governance and Policy Analytics using NLP
Ms. Neha W. Bandabuche, Dr. V. H. Deshmukh, Dr. Y. A. Dhumale
DOI: 10.17148/IJARCCE.2026.15495
Abstract: Artificial Intelligence (AI) is increasingly reshaping how governments function and how societies and economies operate worldwide. It offers significant advantages, such as more efficient financial services, better medical diagnosis, and faster decision-making. At the same time, it raises important concerns, including bias in algorithms, lack of transparency, threats to privacy, and unclear responsibility when AI systems make decisions. For AI to be successfully adopted in the long run, public trust plays a crucial role. However, this trust can be weakened when regulations are unclear or not effectively implemented. This study compares how different regions—such as the United States, the European Union, China, and other emerging AI hubs—approach AI governance. It looks at their regulatory frameworks, ethical guidelines, and institutional practices, using key factors like transparency, fairness, accountability, and stakeholder involvement. The analysis draws on laws, government policies, international standards, and academic research. The results suggest that people are more likely to trust AI systems when policies are clear, transparent, and inclusive. On the other hand, inconsistent or vague regulations can create confusion and slow down the adoption of AI technologies.
Keywords: Artificial Intelligence Governance, Public Trust, AI Regulation, Ethics, Policy Analysis.
Keywords: Artificial Intelligence Governance, Public Trust, AI Regulation, Ethics, Policy Analysis.
Abstract
IoT and Machine Learning Based Air Quality Monitoring and AQI Prediction System
S. Pujitha, M. Rama Krishna [M. Tech, PhD]
DOI: 10.17148/IJARCCE.2026.15496
Abstract: Air pollution has emerged as one of the most critical environmental and public health challenges of the twenty- first century. In India, the majority of urban centres regularly report Air Quality Index (AQI) values in the 'Poor' to 'Hazardous' range, yet real-time, localised air quality data remains scarce and inaccessible due to the high cost and sparse deployment of certified monitoring stations. This paper presents an end-to-end IoT and Machine Learning Based Air Quality Monitoring and AQI Prediction System that addresses these limitations through affordable hardware, cloud connectivity, and intelligent data analysis. The hardware sensing node is built around an ESP32 microcontroller interfaced with five sensors: a DHT11 temperature and humidity sensor, an MQ2 smoke and combustible gas sensor, an MQ7 carbon monoxide sensor, an MQ135 air quality gas sensor, and an optical PM2.5 dust sensor. Readings are displayed locally on a 20×4 I2C LCD and uploaded to a ThingSpeak cloud channel every 15 seconds. The machine learning subsystem employs a Random Forest classifier trained on 1,000 labelled environmental records spanning six AQI categories. The trained model is deployed within a Streamlit web application supporting Manual Input and ThingSpeak Auto modes, generating an AI Environment Report for every prediction. System validation used 82 live readings collected from the prototype hardware on 31 March 2026. The proposed system demonstrates that integration of low-cost IoT sensing, cloud data management, and ensemble machine learning can produce an intelligent air quality monitoring platform suitable for educational institutions and smart city deployment.
Keywords: Internet of Things, Air Quality Index, Arduino UNO, ESP32, MQ Gas Sensors, PM2.5, DHT11, ThingSpeak, Random Forest, Machine Learning, Streamlit, Environmental Monitoring, AQI Prediction, Smart Environment.
Keywords: Internet of Things, Air Quality Index, Arduino UNO, ESP32, MQ Gas Sensors, PM2.5, DHT11, ThingSpeak, Random Forest, Machine Learning, Streamlit, Environmental Monitoring, AQI Prediction, Smart Environment.
Abstract
Automated Soybean Crop Health Evaluation from UAV Images Using Patch-Level CNNs
Saket Bobade, Sangram S. Dandge, Dr. Vaishali H. Deshmukh
DOI: 10.17148/IJARCCE.2026.15497
Abstract: In this paper, we present a simple system to check soybean crop health using images taken from drones (UAVs). In normal farming, checking crop diseases takes a lot of time and effort, and sometimes small disease areas are missed. Because of this, farmers may face loss in yield. So, we tried to make an automatic system which can help in early detection of crop problems.
In our approach, UAV images are not used directly as a whole. Instead, we break each large image into many small parts called patches. This helps the model to focus more on plant areas and less on unwanted background like soil or shadows. For classification, we used a lightweight deep learning model called MobileNetV3-Small. It is not very heavy, so it works faster and can be used even on limited systems.
We also applied some basic data augmentation methods like flipping, rotation, and brightness change. This is done because in real fields, lighting and angles are not always the same. After predicting each patch as healthy or diseased, we combine all results to create a full field health map. This map helps to understand which areas are affected.
The results we got are quite good and show that the method works properly on different UAV images. The system is simple, fast, and can be useful for farmers to take quick decisions. It can also be extended to other crops in future.
Keywords: Soybean crop health, UAV images, drone-based agriculture, patch-level analysis, convolutional neural network, MobileNetV3, plant disease detection, precision agriculture, image classification, deep learning in agriculture
In our approach, UAV images are not used directly as a whole. Instead, we break each large image into many small parts called patches. This helps the model to focus more on plant areas and less on unwanted background like soil or shadows. For classification, we used a lightweight deep learning model called MobileNetV3-Small. It is not very heavy, so it works faster and can be used even on limited systems.
We also applied some basic data augmentation methods like flipping, rotation, and brightness change. This is done because in real fields, lighting and angles are not always the same. After predicting each patch as healthy or diseased, we combine all results to create a full field health map. This map helps to understand which areas are affected.
The results we got are quite good and show that the method works properly on different UAV images. The system is simple, fast, and can be useful for farmers to take quick decisions. It can also be extended to other crops in future.
Keywords: Soybean crop health, UAV images, drone-based agriculture, patch-level analysis, convolutional neural network, MobileNetV3, plant disease detection, precision agriculture, image classification, deep learning in agriculture
Abstract
Enhancing Shared Mobility Through AI-Driven Bike Rental Platforms
Prof. Diksha Bansod, Aaditi Katole, Janvi Aher, Sumit Ghoshal, Jay Ingole
DOI: 10.17148/IJARCCE.2026.15498
Abstract: The increasing demand for efficient urban mobility has led to the rise of digital platforms that provide affordable and flexible transportation options. Bike rental systems, in particular, offer a convenient solution for short- distance travel without the burden of ownership costs. RentBike is designed as a comprehensive web-based platform that connects riders, bike owners, and administrators within a unified system. The backend is developed using Flask to handle server-side operations efficiently, while MongoDB Atlas provides secure and scalable cloud-based data storage. The frontend is built using Tailwind CSS, ensuring a responsive and adaptive interface across different devices such as smartphones and desktops.
The platform integrates multiple functionalities including user authentication, bike search, booking management, payment processing, and review systems. Role-based access control ensures that users, vendors, and administrators have distinct permissions and capabilities within the system. Additionally, the platform incorporates a recommendation mechanism that analyzes user behavior and booking history to suggest suitable bikes. This personalized approach enhances user experience by presenting relevant options based on past interactions. The system operates seamlessly in the background, dynamically adapting to user roles and preferences without requiring explicit intervention.
From a technical perspective, the system is built using a modular architecture where different components interact through well-defined interfaces. Security is maintained using token-based authentication, while optimized database queries improve search performance and response time. The responsive design ensures accessibility across multiple devices, contributing to a smooth user experience. Performance testing indicates that the system can handle high loads efficiently while maintaining reliability and speed. Overall, RentBike demonstrates how modern web technologies can be leveraged to create scalable and intelligent mobility solutions, with potential future enhancements such as real-time tracking, dynamic pricing, and IoT integration.
Keywords: Flask Framework, JWT Authentication, Urban Mobility, RESTful APIs
The platform integrates multiple functionalities including user authentication, bike search, booking management, payment processing, and review systems. Role-based access control ensures that users, vendors, and administrators have distinct permissions and capabilities within the system. Additionally, the platform incorporates a recommendation mechanism that analyzes user behavior and booking history to suggest suitable bikes. This personalized approach enhances user experience by presenting relevant options based on past interactions. The system operates seamlessly in the background, dynamically adapting to user roles and preferences without requiring explicit intervention.
From a technical perspective, the system is built using a modular architecture where different components interact through well-defined interfaces. Security is maintained using token-based authentication, while optimized database queries improve search performance and response time. The responsive design ensures accessibility across multiple devices, contributing to a smooth user experience. Performance testing indicates that the system can handle high loads efficiently while maintaining reliability and speed. Overall, RentBike demonstrates how modern web technologies can be leveraged to create scalable and intelligent mobility solutions, with potential future enhancements such as real-time tracking, dynamic pricing, and IoT integration.
Keywords: Flask Framework, JWT Authentication, Urban Mobility, RESTful APIs
Abstract
Design and Development of Drinking Water Quality Alert System with Voice Announcement for Remote Areas
A. Haresh Kumar, B. Bhartha Kumar, G. Ganga Reddy, Dr. Ome Nerella
DOI: 10.17148/IJARCCE.2026.15499
Abstract: Due to rising levels of water contamination and pollution, ensuring safe drinking water is an essential public health necessity. Traditional techniques for testing the quality of water are labor-intensive, manual, and do not yield data in real time. The development of a drinking water quality monitoring and alert system that allows for ongoing, real-time water quality parameter monitoring is presented in this project. Essential parameters like pH, turbidity, total dissolved solids (TDS), and drinking water temperature are measured by the system using a variety of sensors. These sensors are connected to a Wi-Fi-capable microcontroller, which gathers and sends sensor data to a cloud platform for processing and archiving. When anomalous conditions are found, users or relevant authorities receive immediate alerts that compare the monitored values with accepted safe drinking water limits. This system helps prevent water- borne illnesses, minimizes human labor, and guarantees prompt detection of water contamination. The suggested solution is affordable, scalable, and appropriate for applications involving rural water supplies and smart cities. By making sure that everyone, even those with low literacy levels, receives important warnings, the voice alert feature increases community awareness and response times. This device can be employed in rural and remote areas for long period since the system is made to work 24/7 with little or no human intervention. IoT makes the data to be visually represent to view live data through a dashboard or mobile applications so that authorities and residents can have water quality data with them no matter what location they are in. It logs historical water quality data, used for further analysis and management purposes.
Keywords: IoT for Water Quality Monitoring using ESP32 with pH, TDS, and Turbidity sensors, coupled with an Alert System that is voice controlled and also pushes notificataions on Blynk.
Keywords: IoT for Water Quality Monitoring using ESP32 with pH, TDS, and Turbidity sensors, coupled with an Alert System that is voice controlled and also pushes notificataions on Blynk.
Abstract
DeskMind: A Modular Desktop-Based AI Assistant for Voice and Text Command Processing
Lajari Atul Patil, Aditya Ganesh Poojari, Parth Kishor Rane, Siddhesh Deepak Wagh,Dr. Avinash Sonule
DOI: 10.17148/
Abstract: The increasing demand for intelligent and user- friendly computing systems has led to the development of virtual assistants capable of performing tasks through natural interaction. This paper presents DeskMind, a modular desktop- based AI assistant designed to process user commands using voice or text input and execute tasks efficiently through a scalable architecture. The system integrates speech recognition, command processing, and dynamic skill management to provide a flexible and extensible platform. DeskMind follows a structured design approach where a central assistant module coordinates with a dispatcher to interpret user commands and route them to appropriate functional modules known as skills. These skills are dynamically loaded, allowing the system to be easily extended without modifying the core architecture. The implementation also incorporates multi-threaded execution for improved responsive- ness and logging mechanisms for monitoring and debugging.
Abstract
Hybrid Deep Learning and Transformer-Based Approach for Accurate Email Spam Classification
Dhanashri Shukla, Suyash Shrivastava
DOI: 10.17148/IJARCCE.2026.154100
Abstract: Due to the rapid increase of unsolicited and malevolent messages, the problem of spam detection in email has become a serious issue. In this paper, we propose a comprehensive multi-model framework for spam classification integrating traditional machine learning, deep learning and transformer-based approaches. The system uses feature-based models such as TF-IDF with Logistic Regression and XgBoost, sequence based model like LSTM,CNN,BiLSTM with Attention, transformer-based DistilBERT models Experiments on a labeled email dataset demonstrate that CNN model is most performant with accuracy 98% and F1-score 0.96, BiLSTM with Attention and TF-IDF with XGBoost about 97 percent accurate. Transformer based models also provide competitive results with approx 96% accuracy. The outcome of the models underscores that hybrid and attention-based architectures are critical in improving classification performance, resisting attacks, and adjusting to changing patterns of spam.
Keywords: Email Spam Detection, Machine Learning, Deep Learning, CNN, BiLSTM with Attention, XGBoost, BERT, Text Classification, Natural Language Processing, Hybrid Models
Keywords: Email Spam Detection, Machine Learning, Deep Learning, CNN, BiLSTM with Attention, XGBoost, BERT, Text Classification, Natural Language Processing, Hybrid Models
Abstract
A SYSTEMATIC REVIEW: OF AI-BASED MENTAL HEALTH CHATBOTS: TECHNIQUES, CHALLENGES, AND FUTURE DIRECTIONS
Aman Kumar Sharma, Amit Kumar, Ashish Solanki, Ashish kumar, Aditya Sikarwar, Alok Singh Jadaun
DOI: 10.17148/IJARCCE.2026.154101
Abstract: Mental health disorders such as anxiety, depression, and stress have become increasingly common in recent years, creating a growing need for accessible and effective support systems. Traditional mental healthcare services often face limitations including high costs, limited availability of professionals, and social stigma, which prevent many individuals from seeking timely help. As a result, alternative digital solutions have gained attention in addressing these challenges.
Artificial Intelligence (AI)-based mental health chatbots have emerged as a promising approach to provide immediate, scalable, and anonymous support. These systems use Natural Language Processing (NLP) and machine learning techniques to simulate human-like conversations, enabling users to express their emotions and receive supportive responses. Unlike conventional methods, chatbots offer 24/7 availability and reduce barriers associated with traditional therapy.
This paper presents a systematic review of AI-based mental health chatbots, focusing on their underlying technologies, design approaches, and effectiveness in real-world applications. Various systems such as Woebot, Wysa, and other conversational agents are analyzed to understand their strengths, limitations, and impact on user well-being.
The study further examines key challenges including emotional understanding, ethical concerns, data privacy, and dependency risks associated with AI-driven mental health systems. It also highlights the importance of incorporating empathy, personalization, and safety mechanisms in chatbot design to improve user experience and reliability.
The findings suggest that AI mental health chatbots can serve as effective supplementary tools for early-stage emotional support and mental wellness management. However, they are not a replacement for professional therapy and should be used alongside traditional healthcare systems for optimal outcomes.
Keywords: Artificial Intelligence, Mental Health Chatbots, Natural Language Processing, Emotional Support Systems, Digital Healthcare, Machine Learning, Conversational Agents, Cognitive Behavioral Therapy (CBT)
Artificial Intelligence (AI)-based mental health chatbots have emerged as a promising approach to provide immediate, scalable, and anonymous support. These systems use Natural Language Processing (NLP) and machine learning techniques to simulate human-like conversations, enabling users to express their emotions and receive supportive responses. Unlike conventional methods, chatbots offer 24/7 availability and reduce barriers associated with traditional therapy.
This paper presents a systematic review of AI-based mental health chatbots, focusing on their underlying technologies, design approaches, and effectiveness in real-world applications. Various systems such as Woebot, Wysa, and other conversational agents are analyzed to understand their strengths, limitations, and impact on user well-being.
The study further examines key challenges including emotional understanding, ethical concerns, data privacy, and dependency risks associated with AI-driven mental health systems. It also highlights the importance of incorporating empathy, personalization, and safety mechanisms in chatbot design to improve user experience and reliability.
The findings suggest that AI mental health chatbots can serve as effective supplementary tools for early-stage emotional support and mental wellness management. However, they are not a replacement for professional therapy and should be used alongside traditional healthcare systems for optimal outcomes.
Keywords: Artificial Intelligence, Mental Health Chatbots, Natural Language Processing, Emotional Support Systems, Digital Healthcare, Machine Learning, Conversational Agents, Cognitive Behavioral Therapy (CBT)
Abstract
MINEWATCH: LORA-ENABLED COAL MINE MONITORING WITH RASPBERRY PI PICO
U.Vamsi Krishna, K. Appala Raju
DOI: 10.17148/IJARCCE.2026.154102
Abstract: The mining industry faces significant challenges in ensuring the safety of workers due to the unpredictable nature of underground environments. Hazardous conditions such as methane gas leakage, carbon monoxide presence, high temperature, and water seepage can lead to severe accidents if not detected in time. Traditional monitoring systems rely heavily on manual supervision and wired communication networks, which are prone to failure during emergencies such as explosions or structural collapses.
Keywords: Long-range communication, Works well underground, Low power consumption
Keywords: Long-range communication, Works well underground, Low power consumption
Abstract
SOLARFLARE: AI AND IoT FUSION-BASED SOLAR PANEL PROTECTION SYSTEM
M.Rama Krishna, Sk Shafiulla
DOI: 10.17148/IJARCCE.2026.154103
Abstract: Energy optimization is a crucial aspect of sustainable development, and integrating Artificial Intelligence (AI) and the Internet of Things (IoT) can significantly improve energy efficiency. This project proposes an AI-IoT-based automated power management system that dynamically controls electrical loads based on environmental conditions such as temperature, humidity, light intensity, and object detection. Sensors continuously monitor these parameters and upload the data to a cloud platform. Machine learning algorithms analyze this real-time data to predict optimal energy consumption patterns. Based on predictions, control commands are sent to the Arduino via a serial port, enabling automated switching of appliances. This intelligent system minimizes energy wastage, enhances user comfort, and contributes to cost-effective power utilization.
Keywords: Energy Optimization, Smart Home, Internet of Things (IoT) , Sensor-Based Automation ,Real-Time Monitoring
Keywords: Energy Optimization, Smart Home, Internet of Things (IoT) , Sensor-Based Automation ,Real-Time Monitoring
Abstract
DESIGN AND DEVELOPMENT OF AN ERGONOMIC BULLOCK CART YOKE USING CAD MODELLING AND FINITE ELEMENT ANALYSIS
Atharva Bhedse, Hariom Jat, Anirudh Menon, Raj Patil
DOI: 10.17148/IJARCCE.2026.154104
Abstract: Bullock carts remain an important mode of transportation and agricultural operation in rural regions, where the yoke serves as a key component for transmitting pulling forces from animals to the cart. Conventional yoke designs often suffer from improper load distribution, reduced durability and discomfort to animals due to non-optimized geometry and material selection. The present study focuses on the design and development of an improved bullock cart yoke using computer-aided design and engineering analysis techniques. A three-dimensional model was developed using SolidWorks considering ergonomic and structural requirements. Finite Element Analysis was performed in ANSYS to evaluate stress distribution, deformation behaviour and structural performance under working load conditions. The proposed design demonstrates improved load transfer characteristics, reduced stress concentration and enhanced durability compared to traditional designs. The developed yoke provides better operational efficiency and improved comfort while maintaining cost effectiveness suitable for rural applications. The study highlights the application of modern engineering design tools in optimizing traditional agricultural equipment.
Keywords: Bullock Cart, Yoke Design, Solidworks, Finite Element Analysis, Structural Optimization, Mechanical Design.
Keywords: Bullock Cart, Yoke Design, Solidworks, Finite Element Analysis, Structural Optimization, Mechanical Design.
Abstract
Intelligent Energy Optimization of Electric Vehicles Using Coordinated Regenerative Braking and Advanced Control Strategies
Tulsi D. Kuralkar, Aditya G. Kale, Disha S. Gopatwar, Prof. A. A. Gophane
DOI: 10.17148/IJARCCE.2026.154105
Abstract: Improving energy efficiency remains a critical challenge in the development of electric vehicles (EVs), particularly under real-world driving conditions where braking events significantly influence overall energy consumption. Regenerative braking systems offer an effective solution by converting vehicle kinetic energy into electrical energy during deceleration; however, their performance is strongly dependent on braking force coordination, control strategy design, and power electronic interfaces. Previous studies have investigated braking force coordination for in- wheel motor–driven EVs using electro-hydraulic composite braking systems to enhance stability and energy recovery. Additional research has explored alternative energy recovery mechanisms such as regenerative suspension systems, demonstrating the feasibility of harvesting vibration energy to complement regenerative braking. Advanced modeling and simulation approaches have been proposed to accurately evaluate EV energy efficiency under practical driving conditions braking effectiveness. Fuzzy logic–based approaches, have shown significant potential in optimizing regenerative braking torque distribution, improving braking smoothness, and maximizing energy recovery under system uncertainties. Synchronization control strategies for distributed drive electric vehicles further contribute to coordinated regenerative braking and improved system robustness.
Keywords: Electric vehicles, regenerative braking, energy optimization, intelligent control strategies, braking force coordination, bidirectional power converters, fuzzy logic control, neural control, energy management systems
Keywords: Electric vehicles, regenerative braking, energy optimization, intelligent control strategies, braking force coordination, bidirectional power converters, fuzzy logic control, neural control, energy management systems
Abstract
Smart Waste Management System Using Deep Learning
Mr. M. Rama Krishna P. Rupas
DOI: 10.17148/IJARCCE.2026.154106
Abstract: The rapid increase in waste generation across urban and rural areas demands intelligent, automated solutions for effective waste segregation. This paper presents an AI-Powered Smart Waste Sorting Bin that integrates deep learning, computer vision, and Internet of Things (IoT) technologies to automate the classification of waste into biodegradable and non-biodegradable categories. A camera mounted on a laptop captures images of waste items placed near the bin. An Infrared (IR) sensor connected to a Raspberry Pi Pico microcontroller detects the presence of waste and triggers the image acquisition process. The captured image is then analysed by a trained YOLOv8 (You Only Look Once, version 8) deep learning model, which classifies the waste based on visual features. The classification result is communicated via serial protocol to the Raspberry Pi Pico, which activates servo motors to route the waste into the appropriate bin compartment. A 16×2 LCD display provides real-time feedback to the user. Experimental results confirm that the system achieves reliable waste classification with minimal human intervention, offering a practical and cost-effective solution for smart city waste management.
Keywords: Deep Learning, YOLOv8, Waste Classification, IoT, Raspberry Pi Pico, Smart Waste Management, Computer Vision, Servo Motor
Keywords: Deep Learning, YOLOv8, Waste Classification, IoT, Raspberry Pi Pico, Smart Waste Management, Computer Vision, Servo Motor
Abstract
HERB GUARD AI- AN AI - BASED HERB-DRUG INTERACTION PREDICTION USING NATURAL LANGUAGE PROCESS
Shashipriya Shridhar Hegde, K Sharath
DOI: 10.17148/IJARCCE.2026.154107
Abstract: The growing concurrent use of Ayurvedic and Allopathic medicines has significantly increased the risk of herb–drug interactions (HDIs), many of which remain undetected due to the absence of centralized prediction systems. This paper presents HerbGuard AI, an AI-based system that leverages Natural Language Processing (NLP) to extract interaction-related information from biomedical and traditional Ayurvedic literature. A Knowledge Graph is employed to model and represent structured relationships between herbs, pharmaceutical drugs, and their interaction mechanisms. The proposed system predicts potential interaction risks, thereby improving patient safety and supporting informed clinical decision-making in integrative healthcare. The system is particularly relevant to India's pluralistic health culture, where patients frequently combine modern and traditional medicines without physician supervision. Experimental scenarios demonstrate the system's ability to generate actionable interaction warnings and recommend safe herb–drug combinations.
Keywords: Herb–Drug Interaction, Natural Language Processing, Knowledge Graph, Ayurveda, Artificial Intelligence, Pharmacovigilance, Integrative Medicine, Clinical Decision Support
Keywords: Herb–Drug Interaction, Natural Language Processing, Knowledge Graph, Ayurveda, Artificial Intelligence, Pharmacovigilance, Integrative Medicine, Clinical Decision Support
Abstract
AUTOMATION DETECTION OF STEGANOGRAPHICAL CONTENT USING MACHINE LEARNING
Laxman Bhandarwad, Yogiraj Deshmukh, Nitesh Jadhav, Dr Taware G.G
DOI: 10.17148/IJARCCE.2026.154108
Abstract: The rapid growth of digital communication has significantly increased the use of steganography for secure and covert information exchange. While steganography serves legitimate privacy-preserving purposes, it is also widely exploited for unauthorized communication, cybercrime, and data exfiltration. This research paper presents a novel machine learning–based framework for the automated detection of steganographical content in digital images. The proposed system uses feature extraction, statistical image analysis, and supervised learning techniques to identify hidden data embedded through spatial and frequency-domain steganographic methods.
The study focuses on commonly used embedding techniques such as Least Significant Bit (LSB), transform-domain hiding, and adaptive image embedding. Machine learning classifiers including Support Vector Machine (SVM), Random Forest, Convolutional Neural Network (CNN), and Gradient Boosting are evaluated to improve steganalysis accuracy. Experimental results demonstrate that the CNN-based model achieves superior detection performance in terms of accuracy, precision, recall, F1-score, and robustness against noise and compression.
This work contributes to the field of cybersecurity and digital forensics by providing an intelligent, scalable, and automated solution for detecting concealed information in multimedia files.
Keywords: Steganography, Steganalysis, Machine Learning, CNN, Image Forensics, Cybersecurity, Hidden Data Detection
The study focuses on commonly used embedding techniques such as Least Significant Bit (LSB), transform-domain hiding, and adaptive image embedding. Machine learning classifiers including Support Vector Machine (SVM), Random Forest, Convolutional Neural Network (CNN), and Gradient Boosting are evaluated to improve steganalysis accuracy. Experimental results demonstrate that the CNN-based model achieves superior detection performance in terms of accuracy, precision, recall, F1-score, and robustness against noise and compression.
This work contributes to the field of cybersecurity and digital forensics by providing an intelligent, scalable, and automated solution for detecting concealed information in multimedia files.
Keywords: Steganography, Steganalysis, Machine Learning, CNN, Image Forensics, Cybersecurity, Hidden Data Detection
Abstract
AI-Based Violation Alert System for Helmets
Chandan KL, Dr Kavitha A S, Jeevan Pranav D, Punith HM, Rahul R
DOI: 10.17148/IJARCCE.2026.154109
Abstract: This project presents an AI-based smart helmet system designed for real-time traffic violation detection and automated enforcement. The system captures live video using a helmet-mounted camera and processes it using advanced techniques from Artificial Intelligence and Computer Vision. It employs deep learning models such as YOLO for object detection to identify motorcycles and detect violations like riding without a helmet, followed by number plate recognition using OCR techniques. Upon detecting a violation, the system generates a video clip, extracts the vehicle number, and stores the information in a database for automated fine generation. This approach enhances road safety, reduces manual monitoring, and provides evidence-based enforcement. The proposed system aims to contribute to intelligent transportation systems by offering a scalable, efficient, and real-time solution, although challenges such as environmental conditions, computational requirements, and privacy concerns need to be addressed.
Keywords: Helmet violation detection, YOLOv8, object detection, number plate recognition, OCR, Tesseract, ALPR, computer vision, intelligent transportation, ESP32, deep learning, traffic enforcement.
Keywords: Helmet violation detection, YOLOv8, object detection, number plate recognition, OCR, Tesseract, ALPR, computer vision, intelligent transportation, ESP32, deep learning, traffic enforcement.
Abstract
Accessibility in Fitness Technology: Ensuring Inclusive Digital Health
Rachit Agarwal, Vishesh Vats, Kirtika Saini, Shikhar Chauhan, Ajit Singh, Dr. Uruj Jaleel, Dr. Satish Kumar Soni
DOI: 10.17148/IJARCCE.2026.154110
Abstract: As fitness and health apps proliferate, ensuring they are accessible to all users is critical. We investigate how well modern fitness applications comply with accessibility standards (e.g. WCAG 2.2) and meet the needs of people with visual, auditory, motor, or cognitive impairments. Our research includes a standards review, a literature survey of accessible design (for example, Apple and Android accessibility guidelines), and an evaluation of existing fitness apps. We identify common barriers (e.g. lack of audio descriptions, small touch targets, or missing alt text for images) and compare features across several popular apps. Based on these findings, we outline design and implementation recommendations, including adaptive user interfaces and multimodal feedback. The paper also includes a system architecture diagram illustrating an accessible fitness app. Expected outcomes include a set of guidelines to help developers create more inclusive fitness technologies. This work contributes to reducing health disparities by making fitness tools usable by people with disabilities, thus promoting better physical activity for a wider population.
Keywords: Accessibility, Fitness Technology, Inclusive Design, Mobile Health, Android Application, Usability, Human-Centered Design.
Keywords: Accessibility, Fitness Technology, Inclusive Design, Mobile Health, Android Application, Usability, Human-Centered Design.
Abstract
SMART STOVE SAFETY AND MONITORING SYSTEM
Mrs S. Dharani, Aesteen Abraham, Charumathi. B, Remna R.S
DOI: 10.17148/IJARCCE.2026.154111
Abstract: Gas leakage is a critical safety concern that can lead to severe accidents, including fire hazards, explosions, and loss of human life. Despite the availability of conventional gas regulators and safety devices, many systems rely heavily on human intervention, which increases the risk of delayed responses and negligence. This project proposes a smart and automated gas leakage monitoring and cut-off system using the ESP32 microcontroller to enhance safety and minimize risks.
The system integrates multiple components such as a gas leakage sensor (MQ series), fire sensor, relay module, solenoid valve, servo motor, keypad, and LCD display. The ESP32 acts as the central processing unit, continuously monitoring environmental conditions. When gas leakage or fire is detected, the system automatically triggers safety mechanisms, including shutting off the gas supply using a solenoid valve and mechanically controlling the gas regulator via a servo motor.
Additionally, the built-in Wi-Fi capability of the ESP32 enables remote monitoring and alert notifications, ensuring that users are informed in real time. The LCD display provides immediate feedback on system status, enhancing usability. This intelligent system not only improves safety but also reduces dependency on manual monitoring, making it a reliable solution for modern smart homes and industrial environments.
The system integrates multiple components such as a gas leakage sensor (MQ series), fire sensor, relay module, solenoid valve, servo motor, keypad, and LCD display. The ESP32 acts as the central processing unit, continuously monitoring environmental conditions. When gas leakage or fire is detected, the system automatically triggers safety mechanisms, including shutting off the gas supply using a solenoid valve and mechanically controlling the gas regulator via a servo motor.
Additionally, the built-in Wi-Fi capability of the ESP32 enables remote monitoring and alert notifications, ensuring that users are informed in real time. The LCD display provides immediate feedback on system status, enhancing usability. This intelligent system not only improves safety but also reduces dependency on manual monitoring, making it a reliable solution for modern smart homes and industrial environments.
Abstract
Design And Development of Solar Powered Intelligent Weed Removal system for Sustainable Farming using ESP32
Mrs.Affrose, Padigelawar Hemanth, Vatrapu Sathya Sai Reddy, Routu Girish
DOI: 10.17148/IJARCCE.2026.154112
Abstract: Farming is the source of income for more than half of the Indian population. One of the serious issues in agriculture is the control of weeds growing among the plantation crops. At present weeds are being removed manually by farmers wherever possible, or weed killers/herbicides are being sprayed all over the field to keep them under control. This technique is very inefficient because chemicals are being sprayed on plantation crops also, which leads to, polluting the environment and health problems in humans. To avoid these consequences, a smart weed control system should be deployed. This project presents a solar-powered autonomous robotic system capable of detecting and mechanically removing weeds using edge-based TinyML inference on an ESP32-CAM module. The system captures field images, classifies them using a machine learning model trained via Edge Impulse, and activates a mechanical cutter upon weed detection. The integration of renewable solar energy enhances sustainability and field usability.
Keywords: Weed Detection, TinyML, Edge AI, Real-Time Image Classification, Autonomous Agricultural Robot.
Keywords: Weed Detection, TinyML, Edge AI, Real-Time Image Classification, Autonomous Agricultural Robot.
Abstract
Cloud Guardian: An AI-Assisted Edge-Based AWS Audit and Optimization System
Mr. E. Lingamurthy, Md Majid, P. Mallikarjun, M. Yashwanth
DOI: 10.17148/IJARCCE.2026.154113
Abstract: Cloud providers like Amazon Web Services (AWS) are easy to deploy and operate applications on, but at the same time, they are difficult to keep track of costs and security. It is found that some users have been inadvertently keeping idle virtual machines, unattached storage volumes and outdated snapshots in their accounts, resulting in wastage. Besides, security misconfigure like ports opened to internet can lead to security risks. It is impossible to identify these issues manually since most cloud infrastructure is made up of complex structure of interrelated resources and huge amount of configuration data. This paper proposes a cloud audit and optimization tool Cloud Guardian to address this problem, which is a command line-based tool that performs cloud audit on user’s AWS infrastructure, with the aim of identifying the most expensive and insecure resources. Cloud Guardian performs cloud infrastructure audit by collecting resource data, analyzing resource utilization, and using a rule- based analysis approach to flagged resource inefficiency and insecurity. In addition, Cloud Guardian also calculates a health score of the cloud account to make the analysis results more understandable. To make the analysis results more understandable, an AI advisor module is also built into the tool to explain the problems and suggest the solutions.
Keywords: Cloud Computing, AWS, Cloud Governance, Cost Optimization, Security Analysis, Cloud Audit, Command-Line Interface (CLI), Rule-Based Analysis, AI Advisory Systems, Cloud Resource Monitoring, Health Scoring, Edge Computing, ESP32, Credential Isolation, DevOps Automation, Cloud Security, Infrastructure Optimization.
Keywords: Cloud Computing, AWS, Cloud Governance, Cost Optimization, Security Analysis, Cloud Audit, Command-Line Interface (CLI), Rule-Based Analysis, AI Advisory Systems, Cloud Resource Monitoring, Health Scoring, Edge Computing, ESP32, Credential Isolation, DevOps Automation, Cloud Security, Infrastructure Optimization.
Abstract
A Multi-National Framework for Real-Time Collaborative Cyber Defense: Evaluating FL Architectures and Aggregation Strategies in Heterogeneous NIDS
Abhiram T Sajeev, Adarsh S J, Alen J S, Alfin Jerome, Amila A L
DOI: 10.17148/IJARCCE.2026.154114
Abstract: The rapid evolution of cyber threats necessitates the development of sophisticated Machine Learning (ML) based Network Intrusion Detection Systems (NIDS). However, the efficacy of these systems is often hampered by the sensitive nature of network traffic and stringent privacy regulations, such as GDPR, which prevent organizations and nations from sharing raw data. To address this “privacy-security paradox,” this paper presents a decentralized framework for collaborative threat intelligence utilizing Federated Learning (FL).We simulate a high-stakes multi-national scenario where three distinct nations collaboratively train a global NIDS model while maintaining data sovereignty. The testbed comprises physical nodes representing an aggregation server, a threat actor, and independent nations, with the latter further simulating diverse domestic sectors including Critical Infrastructure (SCADA/IIoT), Financial Services, and Tech Hubs to generate realistic, heterogeneous traffic.
Our research evaluates the performance of various local model architectures, comparing 1D-CNN, DNN, and Autoencoders for detecting complex patterns in network features. Furthermore, we conduct a comparative analysis of aggregation algorithms to mitigate challenges posed by non-IID data. Experimental results demonstrate that the collaborative global model achieves significantly higher detection accuracy than isolated systems. This work provides evidence that Federated Learning is a viable framework for privacy-preserving network security.
Keywords: Federated Learning, NIDS, FedProx, 1D-CNN, Multi-National Cyber Defense, Privacy-Preserving AI, CIC- IDS2017.
Our research evaluates the performance of various local model architectures, comparing 1D-CNN, DNN, and Autoencoders for detecting complex patterns in network features. Furthermore, we conduct a comparative analysis of aggregation algorithms to mitigate challenges posed by non-IID data. Experimental results demonstrate that the collaborative global model achieves significantly higher detection accuracy than isolated systems. This work provides evidence that Federated Learning is a viable framework for privacy-preserving network security.
Keywords: Federated Learning, NIDS, FedProx, 1D-CNN, Multi-National Cyber Defense, Privacy-Preserving AI, CIC- IDS2017.
Abstract
PROSE: Prompt Refinement, Optimization and Semantic Evaluation
Chetan Kokate, Vedant Khadye, Shubham Borate, Chaitanya Kokate
DOI: 10.17148/IJARCCE.2026.154115
Abstract: Prompt optimization has emerged as a critical technique for improving the performance, efficiency, and reliability of Large Language Models (LLMs) without modifying their underlying architecture or parameters. Instead of retraining models, optimized prompts guide the model to generate more accurate, consistent, and context-aware responses. This paper presents a systematic approach to prompt optimization by analyzing prompt structures, refinement strategies, and evaluation techniques. The proposed system focuses on improving response relevance, reducing ambiguity, and minimizing token usage through iterative prompt tuning and rule-based optimization. Experimental observations demonstrate that optimized prompts significantly enhance output quality while reducing computational overhead. The study highlights prompt optimization as a cost-effective and scalable solution for real- world AI applications.
Keywords: Semantic Evaluation, Context-Aware Optimization, Natural Language Processing, Automated Prompt Refinement.
Keywords: Semantic Evaluation, Context-Aware Optimization, Natural Language Processing, Automated Prompt Refinement.
Abstract
RFID-Based Attendance System: Design, Implementation, And Performance Analysis Using ESP32 Microcontroller With RC522 RFID Module And I2C LCD Display
Ms.Charulatha. R.T, Thanish L. S., Sushaanth.M, Milan Prabhu.T, Varun. V.V, Gokul.J
DOI: 10.17148/IJARCCE.2026.154116
Abstract: In most colleges and schools, attendance is still taken using roll calls or written registers. This method takes time and sometimes leads to mistakes, and students may also give proxy attendance.
Because of these problems, an RFID-based attendance system is developed in this work using ESP32, MFRC522 RFID reader, and a 16x2 I2C LCD display.
The system works by scanning RFID cards that operate at 13.56 MHz. Each card has its own unique ID. When a card is placed near the reader, the ID is read using SPI communication and then checked with the IDs already stored in the system. If the ID matches, access is allowed, otherwise it is denied. The result is shown on the LCD screen immediately.
The hardware and software parts were implemented using Embedded C in Arduino IDE. During testing, the system was able to read cards correctly within a small distance of around 0–30 mm. The response time was also quick, usually within about 200 milliseconds.
From the results, the system is simple, reliable, and low in cost, so it can be used easily in real environments. In future, this system can be improved by adding Wi-Fi, storing data in the cloud, and creating a web page to view attendance records.
Keywords: RFID, ESP32, Attendance System, MFRC522, SPI, I2C LCD, Embedded Systems, Arduino IDE
Because of these problems, an RFID-based attendance system is developed in this work using ESP32, MFRC522 RFID reader, and a 16x2 I2C LCD display.
The system works by scanning RFID cards that operate at 13.56 MHz. Each card has its own unique ID. When a card is placed near the reader, the ID is read using SPI communication and then checked with the IDs already stored in the system. If the ID matches, access is allowed, otherwise it is denied. The result is shown on the LCD screen immediately.
The hardware and software parts were implemented using Embedded C in Arduino IDE. During testing, the system was able to read cards correctly within a small distance of around 0–30 mm. The response time was also quick, usually within about 200 milliseconds.
From the results, the system is simple, reliable, and low in cost, so it can be used easily in real environments. In future, this system can be improved by adding Wi-Fi, storing data in the cloud, and creating a web page to view attendance records.
Keywords: RFID, ESP32, Attendance System, MFRC522, SPI, I2C LCD, Embedded Systems, Arduino IDE
Abstract
CEREP: A Graph-Constrained Explainable Reasoning Engine for Multi-Omics Precision Oncology
Sushmita Kundu, Dev Mehta, Charulatha. R. T
DOI: 10.17148/IJARCCE.2026.154117
Abstract: This paper presents the Computational Explainable Reasoning Engine for Precision Oncology (CEREP) for targeted cancer therapeutics. The proposed system integrates deterministic bioinformatics pipelines with structured biological knowledge graphs to optimize algorithmic interpretability, balancing multi-omics integration, causal pathway modelling, hallucination elimination, and clinical auditability. A graph-constrained decoding algorithm processes high- dimensional patient profiles to adaptively regulate Large Language Model (LLM) token generation under strict biological constraints. The framework consists of five synergistic modules: (i) a multi-omics processing layer utilizing nf-core/sarek and quantms for deterministic variant annotation and protein quantification , (ii) a Biolink-compliant knowledge graph constructed via BioCypher for mapping validated biological relationships , (iii) a KG-Trie structural index bounding the LLM search space to established biochemical realities , (iv) a lightweight KG-specialized LLM for constrained multi- hop path extraction, and (v) a Fusion-in-Decoder (FiD) module utilizing a general LLM for inductive clinical narrative synthesis. Experimental evaluation on TCGA and CPTAC breast cancer (BRCA) datasets demonstrates substantial improvements over conventional Retrieval-Augmented Generation (RAG) paradigms, achieving 100% traceable explanation chains, zero biologically invalid reasoning sequences, and highly efficient graph traversal in constant time. The integrated graph-constrained reasoning module achieves state-of-the-art accuracy with zero reasoning hallucination, significantly reducing latency and exhibiting strong zero-shot generalizability to unseen knowledge graphs. Implementation on a FastAPI-Next.js platform with interactive React Flow visualizations ensures transparent, clinical- grade operation. This hybrid framework, combining symbolic graph-theoretic guardrails with vast parametric intelligence, represents a paradigm shift toward intelligent, fully auditable AI systems for next-generation precision oncology.
Keywords: Explainable Artificial Intelligence (XAI), Precision Oncology, Multi-Omics Integration, Graph-Constrained Reasoning, Biological Knowledge Graphs, Large Language Models (LLMs), Proteogenomics, Deterministic Bioinformatics, Mechanistic Explanations, Clinical Decision Support
Keywords: Explainable Artificial Intelligence (XAI), Precision Oncology, Multi-Omics Integration, Graph-Constrained Reasoning, Biological Knowledge Graphs, Large Language Models (LLMs), Proteogenomics, Deterministic Bioinformatics, Mechanistic Explanations, Clinical Decision Support
Abstract
ClashVerse: A Real-Time 1v1 Coding Battle System and Secure Code Execution
Nitin Gupta, Soham Dalvi, Prathmesh Ughade, Mayur Chaudhari, Shilpali Bansu
DOI: 10.17148/IJARCCE.2026.154118
Abstract: Programming practice platforms are widely used to improve problem-solving skills among students and developers. However, most existing platforms focus primarily on individual coding practice and lack real-time interaction between users. This limitation reduces engagement and restricts opportunities for competitive and collaborative learning.
To address this gap, this paper presents ClashVerse, a real-time 1v1 coding battle platform designed to make programming practice more interactive and engaging. The system uses an AI-based matchmaking algorithm to pair users based on their skill level. Participants solve the same programming problem simultaneously using an online code editor, creating a competitive environment.
The submitted solutions are executed securely using the Judge0 API, and performance is evaluated based on correctness and execution efficiency. The platform also updates leaderboards dynamically to reflect user performance. By integrating real-time interaction, gamification, and secure code execution, ClashVerse aims to provide an engaging and competitive platform for coding practice.
Keywords: Competitive Programming, Real-Time Systems, Gamification, AI-Based Matchmaking, Code Execution Platforms
To address this gap, this paper presents ClashVerse, a real-time 1v1 coding battle platform designed to make programming practice more interactive and engaging. The system uses an AI-based matchmaking algorithm to pair users based on their skill level. Participants solve the same programming problem simultaneously using an online code editor, creating a competitive environment.
The submitted solutions are executed securely using the Judge0 API, and performance is evaluated based on correctness and execution efficiency. The platform also updates leaderboards dynamically to reflect user performance. By integrating real-time interaction, gamification, and secure code execution, ClashVerse aims to provide an engaging and competitive platform for coding practice.
Keywords: Competitive Programming, Real-Time Systems, Gamification, AI-Based Matchmaking, Code Execution Platforms
Abstract
E-Commerce With Bidding System
Prof. Bina R. Rewatkar, Priyanshu P. Narayane, Shraddha M. Khodankar, Purva M. Mangrulkar, Anuj A. Kotangale
DOI: 10.17148/IJARCCE.2026.154119
Abstract: This project presents the design and development of an E-commerce platform dedicated to spiritual products, aimed at providing users with a seamless and trustworthy digital marketplace for purchasing items related to spirituality, such as religious pooja items, idols, crystals, and other sacred artifacts. The platform supports standard buying features including product browsing, secure payments, and order management.
A key innovation of this system is the introduction of a special auction-based mechanism for rare and exclusive spiritual products. Selected items are listed under a limited-time bidding system, where users can place competitive offers over a predefined duration (e.g., one week). During this period, the system analyses user interest and demand based on the number of bids and the highest offered price. At the end of the bidding cycle, the product is allocated to the highest bidder, ensuring fair price discovery and maximizing value for both buyers and sellers.
This approach enhances user engagement and creates a dynamic marketplace environment, while maintaining transparency and ethical standards. The platform is designed using modern web technologies to ensure scalability, security, and user-friendly interaction. Overall, the project aims to bridge traditional spirituality with modern digital commerce, offering a unique and efficient solution for spiritual product transactions.
Keywords: E-commerce, Spiritual Products, Auction System, Bidding, Demand Analysis Price Discovery, User Engagement, Digital Marketplace.
A key innovation of this system is the introduction of a special auction-based mechanism for rare and exclusive spiritual products. Selected items are listed under a limited-time bidding system, where users can place competitive offers over a predefined duration (e.g., one week). During this period, the system analyses user interest and demand based on the number of bids and the highest offered price. At the end of the bidding cycle, the product is allocated to the highest bidder, ensuring fair price discovery and maximizing value for both buyers and sellers.
This approach enhances user engagement and creates a dynamic marketplace environment, while maintaining transparency and ethical standards. The platform is designed using modern web technologies to ensure scalability, security, and user-friendly interaction. Overall, the project aims to bridge traditional spirituality with modern digital commerce, offering a unique and efficient solution for spiritual product transactions.
Keywords: E-commerce, Spiritual Products, Auction System, Bidding, Demand Analysis Price Discovery, User Engagement, Digital Marketplace.
Abstract
A Remote Sensing Nano-Satellite Time Control and Communication System
Hamsika M K, Inchara Jain B J, Janhavi M S, Ramya K
DOI: 10.17148/IJARCCE.2026.154120
Abstract: This project focuses on the development of a prototype Nano-satellite system designed for remote sensing applications using an ESP32 microcontroller. The system is capable of handling real-time control and communication tasks, simulating key functionalities of a small satellite platform. It integrates multiple sensors and actuators to monitor environmental conditions and maintain system stability. An accelerometer is used to detect orientation changes, enabling basic attitude control through the activation of two fans positioned on either side of the system. A DHT11 sensor measures temperature, while a gas sensor monitors atmospheric conditions for safety. Additionally, a light sensor is employed to detect ambient light intensity, allowing automatic control of a relay-connected bulb when illumination falls below a predefined threshold. This lighting system can also be manually controlled from a ground station interface. A servo motor is incorporated to facilitate controlled mechanical movement, simulating positioning mechanisms typically found in satellites. For communication, a LoRa module enables long-distance data transmission between the satellite prototype and a ground station. [1-3] An ESP32-based camera module captures images, which are transmitted to a computer for monitoring and analysis. The system is supported by a PC-based ground station that provides real-time visualization of sensor data and allows user interaction for control operations. This setup ensures continuous monitoring and effective management of the satellite’s functions. Overall, the prototype demonstrates how cost-effective IoT components can be utilized to emulate Nano-satellite operations. It serves as a practical platform for research and learning in the fields of remote sensing, embedded systems, and space technology.
Keywords: Nano-satellite prototype, remote sensing, ESP32 microcontroller, real-time embedded systems, attitude control, LoRa communication, wireless telemetry, environmental monitoring, image transmission, MATLAB/Simulink simulation.
Keywords: Nano-satellite prototype, remote sensing, ESP32 microcontroller, real-time embedded systems, attitude control, LoRa communication, wireless telemetry, environmental monitoring, image transmission, MATLAB/Simulink simulation.
Abstract
E-Sports Mobile Application Using Flutter and Firebase
Prof.Diksha Bansod, Aditya A. Langade, Amit S. Kevat, Anurag T. Gajhbiye, Ashvaghosh A. Gaurkhede, Pranav S. Machave
DOI: 10.17148/IJARCCE.2026.154121
Abstract: Nowadays, the eSports industry is growing rapidly, attracting millions of users worldwide. However, users often face difficulty in accessing real-time updates, tournament details, and player statistics in a single platform. To overcome these issues, we propose an eSports Mobile Application using Flutter and Firebase that provides a centralized and user-friendly solution.
The application allows users to view tournaments, track match schedules, explore player and team details, and receive real-time updates. Flutter ensures cross-platform compatibility, while Firebase provides secure authentication and real- time database services. The system improves user engagement by offering a fast, responsive, and visually consistent interface. Testing results show that the application is efficient, scalable, and reliable for real-time usage. This project demonstrates how modern mobile technologies can enhance the accessibility and experience of eSports platforms.
In recent years, the rapid growth of the eSports industry has created a significant demand for efficient and user-friendly digital platforms that provide real-time access to gaming-related information. However, existing systems often suffer from limitations such as lack of integration, delayed updates, inconsistent user interfaces, and platform dependency. To address these challenges, this research presents the design and development of an eSports Mobile Application using Flutter and Firebase, aimed at delivering a centralized and seamless user experience.
The proposed system utilizes Flutter, a cross-platform UI framework, to develop a high-performance mobile application capable of running on both Android and iOS devices using a single codebase. The backend is powered by Firebase, which offers secure authentication, real-time database services, and cloud storage, enabling instant data synchronization and efficient data handling. The application provides key features such as tournament listings, live match schedules, team and player statistics, and real-time notifications, thereby enhancing user engagement and accessibility.
Keywords: E-Sports Application, Flutter, Firebase, Real-Time Updates, Mobile App Development, Cross-Platform, User Interface, Cloud Database.s
The application allows users to view tournaments, track match schedules, explore player and team details, and receive real-time updates. Flutter ensures cross-platform compatibility, while Firebase provides secure authentication and real- time database services. The system improves user engagement by offering a fast, responsive, and visually consistent interface. Testing results show that the application is efficient, scalable, and reliable for real-time usage. This project demonstrates how modern mobile technologies can enhance the accessibility and experience of eSports platforms.
In recent years, the rapid growth of the eSports industry has created a significant demand for efficient and user-friendly digital platforms that provide real-time access to gaming-related information. However, existing systems often suffer from limitations such as lack of integration, delayed updates, inconsistent user interfaces, and platform dependency. To address these challenges, this research presents the design and development of an eSports Mobile Application using Flutter and Firebase, aimed at delivering a centralized and seamless user experience.
The proposed system utilizes Flutter, a cross-platform UI framework, to develop a high-performance mobile application capable of running on both Android and iOS devices using a single codebase. The backend is powered by Firebase, which offers secure authentication, real-time database services, and cloud storage, enabling instant data synchronization and efficient data handling. The application provides key features such as tournament listings, live match schedules, team and player statistics, and real-time notifications, thereby enhancing user engagement and accessibility.
Keywords: E-Sports Application, Flutter, Firebase, Real-Time Updates, Mobile App Development, Cross-Platform, User Interface, Cloud Database.s
Abstract
Design and Development of Intelligent system for Water cleaning Robot using IoT
S. Shiva Sai, Y. Shiva Reddy, Y. Akhil, Mrs K. Rashmi
DOI: 10.17148/IJARCCE.2026.154122
Abstract: Water bodies in urban and rural areas are increasingly polluted due to plastic waste, floating, debris, and other solid pollutants. Manual cleaning of lakes, ponds, and rivers is time consuming, unsafe, and not be efficient for continuous monitoring. This project presents the Design and Development of an Intelligent Water Cleaning Robot using IoT that focuses on automated waste collection from water surfaces. The system uses an Arduino-based control unit integrated with IoT and Bluetooth communication and to operate a floating robot equipped with a rotor-based harvesting mechanism for collecting waste. The robot can be remotely controlled, while sensors and connectivity allow monitoring of operation status. Collected waste is stored onboard, reducing human effort and exposure to polluted water. This system improves the water cleanliness by providing an efficient, low-cost, and intelligent solution for continuous water surface waste removal. This intelligent system reduces human effort, enhances cleaning efficiency, and provides continuous environmental monitoring. The project contributes to sustainable water resource management and supports smart environmental protection systems.
Keywords: Efficient Waste Collection, Cost-Effective Solution, Environmental Protection, Remote Monitoring.
Keywords: Efficient Waste Collection, Cost-Effective Solution, Environmental Protection, Remote Monitoring.
Abstract
ADAPTIVE INTELLIGENT LEARNING SYSTEM USING LARGE LANGUAGE MODELS AND LANGCHAIN
Ms.Vedanti U. Deshmukh, Dr. S. P. Akarte, Dr. G. R. Bamnote
DOI: 10.17148/IJARCCE.2026.154123
Abstract: The rapid evolution of Artificial Intelligence (AI) has revolutionized educational technology, enabling personalized and adaptive learning experiences. This System proposes the development of an Intelligent Educational Agent (IEA) that integrates OpenAI’s GPT-4.1 Turbo through the LangChain framework and employs Retrieval- Augmented Generation (RAG) to deliver contextually rich, dynamic, and learner-centric educational support. Unlike traditional chatbots that depend on single embeddings and static datasets, the proposed IEA adopts a multi-embedding hybrid retrieval mechanism, facilitating semantic understanding across diverse educational resources, including textbooks, lecture materials, and research notes.
A student difficulty tracking module is incorporated to monitor learner performance through metrics such as response accuracy, completion time, and hint usage. These insights enable the agent to classify learners into beginner, intermediate, and advanced categories, generating personalized responses, adaptive quizzes, and progressive content suited to each learner’s cognitive level. Furthermore, the system employs feedback-driven continuous improvement, refining retrieval and content generation strategies based on user interaction patterns.
Experimental evaluations demonstrate that the IEA enhances contextual understanding, engagement, and learning retention by offering customized learning pathways and real-time adaptation. This System establishes a novel framework for multi-source, AI-driven educational assistance, bridging the gap between static content delivery and intelligent, adaptive pedagogy. Future extensions may include multimodal learning integration, multilingual capabilities, and advanced pedagogical modelling, marking a significant contribution to the next generation of intelligent educational technologies.
Keywords: Intelligent Educational Agent, Lang Chain, Large Language Models, Retrieval-Augmented Generation (RAG), GPT-4.1Turbo, Adaptive Learning, Personalized Education, Hybrid Retrieval, Semantic Search, Multi- Embedding.
A student difficulty tracking module is incorporated to monitor learner performance through metrics such as response accuracy, completion time, and hint usage. These insights enable the agent to classify learners into beginner, intermediate, and advanced categories, generating personalized responses, adaptive quizzes, and progressive content suited to each learner’s cognitive level. Furthermore, the system employs feedback-driven continuous improvement, refining retrieval and content generation strategies based on user interaction patterns.
Experimental evaluations demonstrate that the IEA enhances contextual understanding, engagement, and learning retention by offering customized learning pathways and real-time adaptation. This System establishes a novel framework for multi-source, AI-driven educational assistance, bridging the gap between static content delivery and intelligent, adaptive pedagogy. Future extensions may include multimodal learning integration, multilingual capabilities, and advanced pedagogical modelling, marking a significant contribution to the next generation of intelligent educational technologies.
Keywords: Intelligent Educational Agent, Lang Chain, Large Language Models, Retrieval-Augmented Generation (RAG), GPT-4.1Turbo, Adaptive Learning, Personalized Education, Hybrid Retrieval, Semantic Search, Multi- Embedding.
Abstract
Availability-Aware Multimodal Deep Learning for Breast Cancer Diagnosis with Missing Modalities
Indu P. K, Dr. G Beni, Dr. D Rene Dev
DOI: 10.17148/IJARCCE.2026.154124
Abstract: The diagnosis of breast cancer frequently gains from multimodal imaging; however, comprehensive multimodal data are commonly inaccessible in standard clinical practice due to workflow challenges, budget restrictions, and uneven access to imaging resources. Many current multimodal deep learning models have restricted clinical use since they presume that all imaging modalities are accessible during inference. To overcome this constraint, we introduce a multimodal deep learning framework that is aware of missing modalities, incorporating modality availability modeling alongside reliability-informed decision support for breast cancer detection. The system utilizes modality-specific ResNet- 101 encoders for both mammography and ultrasound, along with a fusion module that is aware of availability and that dynamically modifies the contribution of each modality based on its presence. A parallel reliability estimation head forecasts diagnostic confidence, allowing uncertainty-informed clinical recommendations instead of imposing binary choices in unclear situations. A training approach consisting of two stages with random modality dropout is employed to enhance resilience in cases where one or more imaging modalities are absent. Experimental findings indicate that the suggested framework exhibits consistent performance in the presence of missing modalities, while also attaining robust diagnostic discrimination, achieving an AUC of as high as 0.98. In contrast to unimodal models, the multimodal framework generated more accurately calibrated predictions, exhibiting a significantly reduced Expected Calibration Error. Reliability-stratified analysis showed that predictions with high confidence were significantly more accurate than those with low confidence, reinforcing the clinical importance of the suggested reliability score. The proposed framework enhances practical, uncertainty-aware multimodal decision support in real clinical environments by explicitly modeling modality availability and diagnostic confidence.
Keywords: Breast cancer; Multimodal imaging; Missing-modality learning; Availability-aware fusion; Diagnostic reliability; Clinical decision support; Imaging informatics; Deep learning.
Keywords: Breast cancer; Multimodal imaging; Missing-modality learning; Availability-aware fusion; Diagnostic reliability; Clinical decision support; Imaging informatics; Deep learning.
Abstract
A Hybrid Signal-Processing and Neural Network Pipeline for Low-Latency Acoustic Event Detection at the Edge
Jatin Verma, Abhishek Jatla, Thanuja Reddy, Ms. Charulatha R.T
DOI: 10.17148/IJARCCE.2026.154125
Abstract: This paper presents an edge-computed AI framework that detects a specific acoustic event — a human handclap — and actuates a wireless IoT device in response, without touching a cloud server at any point. Audio is captured through a professional condenser microphone, streamed into a Python processing pipeline via a non-blocking callback queue, and subjected to two lightweight deterministic gates before any neural computation occurs. The first gate discards frame whose peak amplitude falls below a fixed noise floor, keeping the classifier idle during silence. The second gate enforces a post-detection bypass window, preventing room echoes from generating spurious follow-on triggers. Frames that clear both gates pass through a two-stage neural pipeline: Google's YAMNet model extracts 1024- dimensional acoustic embeddings, and a purpose-trained Keras Sequential classifier maps those embeddings to a binary confidence score. When confidence exceeds 0.28, the system sends a UDP packet over the local Wi-Fi network to an ESP32 microcontroller, whose on-board LED toggles as hardware confirmation of successful end-to-end delivery. The entire loop — from live audio capture, through AI inference, to physical actuation — runs within a sub-second latency budget on commodity CPU hardware, with no internet dependency. The result is a privacy-preserving, network- independent architecture that generalises to any latency-critical acoustic IoT trigger application.
Keywords: Edge Computing; Acoustic Event Detection; YAMNet; Transfer Learning; Keras; ESP32; Real-Time Signal Processing; Deterministic Frame-Skipping; IoT Actuation.
Keywords: Edge Computing; Acoustic Event Detection; YAMNet; Transfer Learning; Keras; ESP32; Real-Time Signal Processing; Deterministic Frame-Skipping; IoT Actuation.
Abstract
Multilingual AI-Based Voice-Controlled Robotic System Using Distributed Architecture
Mrs. V. Divya Vani, Dr. G. Anand Kumar, K. Dharan, Ch. Tharun
DOI: 10.17148/IJARCCE.2026.154126
Abstract: Human–robot interaction is evolving rapidly with the convergence of artificial intelligence, cloud computing, and embedded systems. This paper presents the design and implementation of Astra, a multilingual AI-
driven voice-controlled robotic system that operates through a distributed architecture. High-level natural language intelligence runs on a laptop while real-time motor control is managed by an ESP32 microcontroller. The system supports three languages—English, Hindi, and Telugu—enabling broad accessibility across India’s linguistically diverse population. Audio input is captured via a wired microphone and transcribed by Sarvam AI, a cloud-based speech recognition service optimized for Indian languages.
The transcribed text is forwarded to a GPT-4o-mini large language model via OpenRouter, which classifies the input as either a movement command or a general conversational query and generates a structured JSON response. Movement commands are transmitted from the laptop to the ESP32 over Wi-Fi using the HTTP protocol, while conversational answers are spoken aloud via gTTS. A soft wake-word mechanism (“Astra”) enhances usability without strict keyword dependency.
Experimental evaluation demonstrates an average speech recognition accuracy of 85 % across all three languages, end- to-end command latency under 2 s, and robust motor control with no packet loss over a local Wi-Fi
Keyword: Human–Robot Interaction, Multilingual Speech Recognition, Sarvam AI, GPT-4o-mini, ESP32, Distributed AI Architecture, Wi-Fi HTTP Control, Natural Language Understanding, Voice-Controlled Robot, IoT.
driven voice-controlled robotic system that operates through a distributed architecture. High-level natural language intelligence runs on a laptop while real-time motor control is managed by an ESP32 microcontroller. The system supports three languages—English, Hindi, and Telugu—enabling broad accessibility across India’s linguistically diverse population. Audio input is captured via a wired microphone and transcribed by Sarvam AI, a cloud-based speech recognition service optimized for Indian languages.
The transcribed text is forwarded to a GPT-4o-mini large language model via OpenRouter, which classifies the input as either a movement command or a general conversational query and generates a structured JSON response. Movement commands are transmitted from the laptop to the ESP32 over Wi-Fi using the HTTP protocol, while conversational answers are spoken aloud via gTTS. A soft wake-word mechanism (“Astra”) enhances usability without strict keyword dependency.
Experimental evaluation demonstrates an average speech recognition accuracy of 85 % across all three languages, end- to-end command latency under 2 s, and robust motor control with no packet loss over a local Wi-Fi
Keyword: Human–Robot Interaction, Multilingual Speech Recognition, Sarvam AI, GPT-4o-mini, ESP32, Distributed AI Architecture, Wi-Fi HTTP Control, Natural Language Understanding, Voice-Controlled Robot, IoT.
Abstract
Enhancing Social Network Prediction in Graph Neural Networks Using Graph Theoretical Approaches: Social Network Analysis on The Facebook Ego Dataset
Pharsana Parveen M, Sr. Stanis Arul Mary A
DOI: 10.17148/IJARCCE.2026.154127
Abstract: This paper focuses on social network analysis and prediction by incorporating Graph theoretical concepts into the architecture of Graph Neural Networks (GNNs), with a focus on Facebook Ego Dataset. We propose an algorithm to leverage graph theoretical principles such as graph coloring, mutual friends, and weighted edge traversal to enhance link prediction tasks in Graph Neural Network. This approach optimizes Graph Neural Network’s performance by capturing local and global structural patterns within social networks. By using graph theoretical technique-based algorithm, the developed model aims on improved accuracy and diversity of friend recommendation in social networks. The study helps us to understand how integrating graph coloring helps in enhancing node embedding. The proposed algorithm generates improved social predictions with higher accuracy and meaningful insights. The results accentuate the importance of applying classical graph theoretical concepts with recent deep learning techniques, providing an efficient framework for social network prediction and analysis.
Keywords: Graph Neural Network, Product Recommendation System, Node Embeddings. Graph Coloring, User Product Graph.
Keywords: Graph Neural Network, Product Recommendation System, Node Embeddings. Graph Coloring, User Product Graph.
Abstract
IOT Based SOS Smart Helmet with Automatic Accident Detection and Real – Time Location Alert System
Akash Kumar, Sudharsan J, Jeron Sam Varghese, Sanjith TS, Hariharan N, Ms. Charulatha R.T
DOI: 10.17148/IJARCCE.2026.154128
Abstract: Every year, thousands of riders die in road accidents — not always because of the crash itself, but because help arrived too late. A helmet that could automatically call for help the moment something goes wrong could change that. That's exactly what this project is about.
The Smart SOS Helmet is an IoT-based system built into a regular motorcycle helmet. It watches for accidents in real time, and the moment it detects a serious impact, it sends an emergency SMS with the rider's exact location to up to three contacts — all without the rider having to do a thing.
How it works At the heart of the system is an MPU-6050 sensor that constantly reads acceleration and tilt data across three axes. The moment the impact crosses a threshold — tuned specifically to ignore normal road vibrations — a 10-second countdown begins. A buzzer sounds, giving the rider a chance to cancel if it was a false alarm. If nobody cancels, the system assumes the rider is unconscious or unable to respond, and the SIM800L GSM module fires off an SMS. That message includes the rider's name, the time of the incident, and a live Google Maps link pulled from the NEO-6M GPS module. The whole thing runs on an Arduino Nano or an ESP32, tucked into the helmet with a small rechargeable Li-ion battery. There's also an IR sensor inside the helmet that checks whether it's actually being worn. If it isn't, the bike simply won't start — a quiet but effective push toward making helmets non-negotiable.
How it performed Testing across both simulated crashes and real riding conditions showed solid results: accident detection accuracy above 92%, GPS location within a 5-metre radius, and the SMS reaching contacts in under 8 seconds from impact. False alerts were kept low through careful threshold calibration combined with that confirmation delay window.
Why it matters The system is affordable, lightweight, and doesn't require the rider to carry anything extra or remember to enable anything. It fills a gap that existing road safety infrastructure largely ignores — the critical window between when an accident happens and when someone actually finds out. For riders who end up unconscious on an empty road, that window is often the difference between life and death.
Down the road, the system can be expanded with cloud logging via Firebase or ThingSpeak, a companion app for live tracking, ML-based fall detection, and eventually integration into smart city road safety networks. But even in its current form, it's a genuinely practical tool that could save lives right now.
The Smart SOS Helmet is an IoT-based system built into a regular motorcycle helmet. It watches for accidents in real time, and the moment it detects a serious impact, it sends an emergency SMS with the rider's exact location to up to three contacts — all without the rider having to do a thing.
How it works At the heart of the system is an MPU-6050 sensor that constantly reads acceleration and tilt data across three axes. The moment the impact crosses a threshold — tuned specifically to ignore normal road vibrations — a 10-second countdown begins. A buzzer sounds, giving the rider a chance to cancel if it was a false alarm. If nobody cancels, the system assumes the rider is unconscious or unable to respond, and the SIM800L GSM module fires off an SMS. That message includes the rider's name, the time of the incident, and a live Google Maps link pulled from the NEO-6M GPS module. The whole thing runs on an Arduino Nano or an ESP32, tucked into the helmet with a small rechargeable Li-ion battery. There's also an IR sensor inside the helmet that checks whether it's actually being worn. If it isn't, the bike simply won't start — a quiet but effective push toward making helmets non-negotiable.
How it performed Testing across both simulated crashes and real riding conditions showed solid results: accident detection accuracy above 92%, GPS location within a 5-metre radius, and the SMS reaching contacts in under 8 seconds from impact. False alerts were kept low through careful threshold calibration combined with that confirmation delay window.
Why it matters The system is affordable, lightweight, and doesn't require the rider to carry anything extra or remember to enable anything. It fills a gap that existing road safety infrastructure largely ignores — the critical window between when an accident happens and when someone actually finds out. For riders who end up unconscious on an empty road, that window is often the difference between life and death.
Down the road, the system can be expanded with cloud logging via Firebase or ThingSpeak, a companion app for live tracking, ML-based fall detection, and eventually integration into smart city road safety networks. But even in its current form, it's a genuinely practical tool that could save lives right now.
Abstract
LUMEX: A Web-Based Image Enhancement Studio Using Python, Flask, and Pillow
Prataparao Charith, K.Varun, Yash Kumar, Dr. H. Mary Shyni
DOI: 10.17148/IJARCCE.2026.154129
Abstract: Digital image processing is a critical component of modern software architecture, yet existing solutions frequently oscillate between prohibitive complexity for casual users and expensive proprietary licensing. This paper introduces LUMEX, a lightweight, browser-based studio for image enhancement developed using Python-Flask (≥2.3.0) with Pillow (≥10.0.0) and NumPy (≥1.24.0). The proposed system integrates a modular pipeline encompassing four tone controls (brightness, contrast, saturation, sharpness), eleven specialized filters, and high-fidelity geometric transformations utilizing LANCZOS resampling. The frontend, implemented as a single-page application with a dark-themed interface, supports drag-and-drop upload, real-time parameter adjustment, and a three-mode image viewer (Original / Enhanced / Split). Architecturally, LUMEX exposes four stateless REST endpoints over HTTP. Experimental results demonstrate sub-300 ms processing latency for HD images on commodity hardware, with JPEG output quality configurable from 10 to 100 (default: 92). By utilizing a fully self-hosted, open-source stack, LUMEX delivers a secure and professional-grade alternative to cloud-based editors that compromise user data privacy.
Keywords: Digital image processing, Flask, Image enhancement, LANCZOS resampling, Pillow, Python, REST API, Web application.
Keywords: Digital image processing, Flask, Image enhancement, LANCZOS resampling, Pillow, Python, REST API, Web application.
Abstract
Gesture Talk
Tejas Sawant, Gaurav Singh, Shubham Tiwari, Aman Verma, Dipali Shende
DOI: 10.17148/IJARCCE.2026.154131
Abstract: Communication is one of the most basic needs for humans, but millions of people who are deaf or hard of hearing struggle to get information and connect with the world around them. The tools available to help them are often limited, providing only pictures that don’t show the full, natural way sign language works. Many platforms also don’t respond quickly or allow people to use their voice, making them hard to use for those with different needs. This paper introduces a web application that helps bridge the communication gap for the deaf and hard-of-hearing community by converting text and speech into real-time, animated Indian Sign Language (ISL). The app lets users type or speak naturally, and then turns that into smooth, animated signs. It uses speech recognition to turn spoken words into text, which is then mapped to an animation system that shows the signs accurately. Instead of using still images, the system creates smooth animations that help people understand better and stay engaged. The web-based design works on any device without needing extra software. By combining speech recognition, language processing, and animation all in one place, the app offers a simple and inclusive way for the deaf community to communicate more easily.
Keywords: Indian Sign Language, speech recognition, real-time conversion, web application, deaf accessibility, animation rendering.
Keywords: Indian Sign Language, speech recognition, real-time conversion, web application, deaf accessibility, animation rendering.
Abstract
Course Cloud: AI-Based Video Description Generator for Smart E- Learning
Shivam Giri, Shivani Kashyap, Shobhit Sharma, Udit Tyagi, Usha Kumari, Dr. Uruj Jaleel, Dr. Satish Soni
DOI: 10.17148/IJARCCE.2026.154132
Abstract: Education is undergoing a significant transformation due to the rise of digital platforms and online learning systems. With the growing dependence on video-based educational content, learners often face difficulties in efficiently searching, understanding, and navigating course materials. Conventional approaches to content description, such as manual tagging and summarization, are not only time-intensive but also inconsistent and difficult to scale. In recent years, Artificial Intelligence (AI) and Machine Learning (ML) have provided effective solutions for automating content analysis and enhancing accessibility within e-learning environments. This paper presents a detailed review of AI- driven techniques for automatic video description generation, with a particular focus on their application in advanced e- learning platforms like Course Cloud.
The study examines a variety of machine learning and deep learning methods, including Natural Language Processing (NLP), computer vision, and sequential models such as Recurrent Neural Networks (RNNs), Long Short-Term Memory (LSTM) networks, and transformer-based architectures. These approaches are analyzed based on factors such as accuracy, scalability, contextual comprehension, and their capability to produce coherent and human-like descriptions from video data.
Additionally, the paper highlights the broader impact of AI in education, including personalized learning experiences, improved content discoverability, accessibility for differently-abled users, and intelligent course recommendation systems. Emerging trends such as real-time video processing and explainable AI in education are also discussed. To address the challenges posed by the increasing volume of video-based learning content, this paper introduces Course Cloud—an AI-powered system designed to automatically generate descriptive summaries for educational videos. By integrating deep learning techniques, NLP, and computer vision, the system effectively converts video content into meaningful textual descriptions.
The proposed solution significantly improves the efficiency and effectiveness of digital learning platforms by enabling better navigation, enhancing accessibility, and supporting adaptive learning. Experimental results indicate that AI-based video description systems can greatly enhance the overall learning experience in modern e-learning environments.
The study examines a variety of machine learning and deep learning methods, including Natural Language Processing (NLP), computer vision, and sequential models such as Recurrent Neural Networks (RNNs), Long Short-Term Memory (LSTM) networks, and transformer-based architectures. These approaches are analyzed based on factors such as accuracy, scalability, contextual comprehension, and their capability to produce coherent and human-like descriptions from video data.
Additionally, the paper highlights the broader impact of AI in education, including personalized learning experiences, improved content discoverability, accessibility for differently-abled users, and intelligent course recommendation systems. Emerging trends such as real-time video processing and explainable AI in education are also discussed. To address the challenges posed by the increasing volume of video-based learning content, this paper introduces Course Cloud—an AI-powered system designed to automatically generate descriptive summaries for educational videos. By integrating deep learning techniques, NLP, and computer vision, the system effectively converts video content into meaningful textual descriptions.
The proposed solution significantly improves the efficiency and effectiveness of digital learning platforms by enabling better navigation, enhancing accessibility, and supporting adaptive learning. Experimental results indicate that AI-based video description systems can greatly enhance the overall learning experience in modern e-learning environments.
Abstract
“Explainable Bone Tumor Diagnosis Using Deep CNNs and Language Models”
Prof. Amit Meshram, Vishal Suresh Nemade, Abhishek Sudam Pawar
DOI: 10.17148/IJARCCE.2026.154133
Abstract: Bone cancer is a critical medical condition that requires early and accurate diagnosis to improve patient outcomes. Traditional diagnostic approaches based on manual interpretation of X-ray images are often limited by inter- observer variability and the scarcity of expert radiologists, particularly in resource-constrained settings. To address these challenges, this paper proposes an explainable deep learning framework for automated bone tumor classification using X-ray images.
The proposed system leverages a pre-trained Convolutional Neural Network (CNN) to accurately detect and classify bone tumors into multiple categories. To enhance model interpretability, Gradient-weighted Class Activation Mapping (Grad-CAM) is employed to generate visual explanations by highlighting the most relevant regions in the input images that contribute to the model’s predictions. Furthermore, an integrated Large Language Model (LLM) module generates human-readable diagnostic explanations, summarizing tumor characteristics, predicted severity, and potential clinical insights.
The framework is evaluated on benchmark medical imaging datasets, demonstrating superior performance compared to conventional machine learning and deep learning baselines. Experimental results show significant improvements in classification accuracy, precision, recall, and F1-score, while also reducing training time. The combination of high predictive performance and enhanced interpretability makes the proposed system a reliable decision-support tool for healthcare professionals.
This work contributes toward the development of transparent, efficient, and accessible AI-driven diagnostic systems, with strong potential for real-world deployment in clinical and low-resource environments.
Keywords: Bone cancer; Deep learning; Convolutional Neural Network (CNN); X-ray classification; Grad-CAM; Large Language Model (LLM).
The proposed system leverages a pre-trained Convolutional Neural Network (CNN) to accurately detect and classify bone tumors into multiple categories. To enhance model interpretability, Gradient-weighted Class Activation Mapping (Grad-CAM) is employed to generate visual explanations by highlighting the most relevant regions in the input images that contribute to the model’s predictions. Furthermore, an integrated Large Language Model (LLM) module generates human-readable diagnostic explanations, summarizing tumor characteristics, predicted severity, and potential clinical insights.
The framework is evaluated on benchmark medical imaging datasets, demonstrating superior performance compared to conventional machine learning and deep learning baselines. Experimental results show significant improvements in classification accuracy, precision, recall, and F1-score, while also reducing training time. The combination of high predictive performance and enhanced interpretability makes the proposed system a reliable decision-support tool for healthcare professionals.
This work contributes toward the development of transparent, efficient, and accessible AI-driven diagnostic systems, with strong potential for real-world deployment in clinical and low-resource environments.
Keywords: Bone cancer; Deep learning; Convolutional Neural Network (CNN); X-ray classification; Grad-CAM; Large Language Model (LLM).
Abstract
Design and development of Sewage Drain Monitoring and alert system using IoT and LoRa Technology
P. Shanmukha Kumar, T. Srujan, Mohd. Faraz
DOI: 10.17148/IJARCCE.2026.154134
Abstract: Urban sewage systems frequently face issues such as blockage and overflow, resulting in unhygienic conditions, health hazards, and damage to public infrastructure. Manual inspection of sewage drains is unsafe, time- consuming, and inefficient. To address these challenges, this project focuses on the design and development of a sewage drain monitoring and alert system using IoT and LoRa technology. The system continuously monitors sewage water levels using sensors and detects abnormal conditions at an early stage.
The sensed data is processed by a microcontroller and transmitted over long distances using LoRa communication, which offers low power consumption and wide coverage. A LoRa gateway forwards the data to a cloud platform for real-time monitoring and alert generation. When overflow or blockage is detected, instant notifications are sent to concerned authorities. The proposed system is cost-effective, reliable, and scalable, making it suitable for smart city sewage management and public health improvement.
Keywords: Sewage Monitoring, IoT, LoRa, Gas Sensor, Level Sensor, Smart Drainage, Alert System
The sensed data is processed by a microcontroller and transmitted over long distances using LoRa communication, which offers low power consumption and wide coverage. A LoRa gateway forwards the data to a cloud platform for real-time monitoring and alert generation. When overflow or blockage is detected, instant notifications are sent to concerned authorities. The proposed system is cost-effective, reliable, and scalable, making it suitable for smart city sewage management and public health improvement.
Keywords: Sewage Monitoring, IoT, LoRa, Gas Sensor, Level Sensor, Smart Drainage, Alert System
Abstract
Real-Time Face Emotion, Hand Gesture Recognition and Voice Conversion System for Deaf and Dumb
Nehan Naveen, Akshay Raj, Gopika C, Sibi D
DOI: 10.17148/IJARCCE.2026.154135
Abstract: This research addresses the profound communication barrier faced by hundreds of millions of individuals globally who rely on sign language as their primary means of expression. Traditional assistive technologies have often relied on expensive and cumbersome wearable hardware, which can be intrusive and impractical for daily use. We present a non-intrusive, real-time assistive system that utilizes a standard webcam and advanced computer vision to bridge this gap. Our approach goes beyond simple word-for-word translation by integrating a dual-layered neural network architecture that simultaneously tracks complex hand movements and analyses subtle facial micro-expressions. By capturing these emotional cues, the system moves beyond robotic, monotone outputs to generate synthesized speech that reflects the user’s true intent—whether it be urgency, joy, or concern. Experimental results demonstrate high classification accuracy across a diverse vocabulary of signs with minimal processing delay, even in varied environmental conditions. This study offers a human-centric solution designed to restore the personality of the speaker and foster more natural, inclusive interactions in critical settings such as healthcare, education, and public services
Keywords: Artificial Intelligence, Assistive Technology, Computer Vision, Deep Learning, Emotion Detection, Gesture Recognition, Human-Computer Interaction, Sign Language Translation, Speech Synthesis.
Keywords: Artificial Intelligence, Assistive Technology, Computer Vision, Deep Learning, Emotion Detection, Gesture Recognition, Human-Computer Interaction, Sign Language Translation, Speech Synthesis.
Abstract
Edutrack Smart System
Varda Khan, Krishikka Panchal, Robin Nadar
DOI: 10.17148/IJARCCE.2026.154136
Abstract: The rapid advancement of smart educational technologies has enabled the development of intelligent classroom environments that enhance both administrative efficiency and learning quality. This paper proposes an integrated smart classroom system that combines embedded hardware with computer vision-based analytics to automate attendance and measure student engagement. The system utilizes an Arduino Uno microcontroller connected to an LCD display for real- time timetable visualization, while a Python-based web application manages student data and lecture sessions. A key contribution of this work is the introduction of a concentration-based attendance mechanism, where students are marked present only if they maintain visual engagement for at least 75% of the lecture duration. A vision module continuously processes video frames using face detection and recognition techniques to track student presence and attention over time. Experimental evaluation demonstrates that the system achieves high accuracy in attendance tracking while effectively reducing proxy attendance. The integration of real-time analytics and visualization further enhances faculty decision- making. This system provides a scalable and cost-effective solution for modern smart classrooms. This integration of IoT-enabled hardware and AI-driven analytics provides a unified and practical solution for modern smart classroom environments.
Keywords: Smart Classroom, Automated Attendance System, Face Recognition, Computer Vision, Student Engagement Analysis, Concentration-Based Attendance, Internet of Things, IoT, Arduino Uno, Real-Time Monitoring, Human- Computer Interaction.
Keywords: Smart Classroom, Automated Attendance System, Face Recognition, Computer Vision, Student Engagement Analysis, Concentration-Based Attendance, Internet of Things, IoT, Arduino Uno, Real-Time Monitoring, Human- Computer Interaction.
Abstract
An Intelligent Video Surveillance System for Multi-Threat Detection and Real-Time Alerting
Aniket Ligam, Avadhoot Katte, Atul Shelar, Niharika Dasari, Rakesh Suryawanshi
DOI: 10.17148/IJARCCE.2026.154137
Abstract: Traditional CCTV surveillance systems rely heavily on continuous human monitoring, which is inefficient, error-prone, and unsuitable for large-scale deployment, as operators may miss critical events due to fatigue, overlapping camera feeds, poor lighting condi-tions, and the difficulty of tracking multiple screens simultaneously. To overcome these limitations, this paper presents an Intelligent Video Surveillance System (IVSS) capable of detecting multiple safety and security threats in real time. The proposed system integrates object detection, weapon detection, fire and smoke detection, fall detection, crash detection, and face recognition with watchlist matching into a unified pipeline. A YOLO-based model is employed for fast and accurate detection of persons and suspicious objects, while dedicated modules analyze fire patterns, abnormal human posture, sudden motion changes, and identity verification against a watchlist database. The system supports both live camera streams and recorded video input, making it flexible for various surveillance scenarios. To improve reliability and reduce false alarms, confidence-based filtering and temporal consistency checks across consecutive frames are applied before generating alerts. Detected incidents are stored as evidence in the form of annotated frames or video clips for further analysis. The modular architecture enables flexible deployment by allowing individual detection components to operate independently or as part of an integrated system. The proposed IVSS is well-suited for security-sensitive and safety-critical environments, providing en-hanced monitoring efficiency, reduced human workload, and faster response to potential threats.
Keywords: Intelligent Surveillance, YOLO, Face Recognition, Watchlist Matching, Fall Detection, Fire Detection, Real-Time Monitoring, Object Detection
Keywords: Intelligent Surveillance, YOLO, Face Recognition, Watchlist Matching, Fall Detection, Fire Detection, Real-Time Monitoring, Object Detection
Abstract
Fuzzy Inference Systems for Optimized Drug Dosing in Heart Failure Management
Varsha*, Prof. Diwari Lal
DOI: 10.17148/IJARCCE.2026.154138
Abstract: The proposed fuzzy inference system uses a number of key clinical variables (i.e. systolic blood pressure, estimated glomerular filtration rate, serum potassium, symptoms of congestive heart failure, overall symptom severity and Nt-Pro BNP burden) to provide an optimal drug dosing solution for patients suffering from heart failure in a logical manner that is understandable to the clinician. It takes various forms of input data and normalizes it; then uses fuzzy logic mapping techniques to map the input data to fuzzy set membership values and apply a Mamdani type fuzzy logic rule base with defuzzification using the centroid method to determine recommended intensity levels for administration of three types of drugs used to treat heart failure (loop diuretics, Angiotensin Receptor-Neprilysin Inhibitors and Mineralocorticoid Receptor Antagonists). Additionally, this system includes a "safety filter" which prevents the generation of potentially unsafe dose recommendations by preventing recommendations that violate certain critical thresholds associated with renal function, elevated serum potassium or hypotension. Thus, although no other studies have been identified that use fuzzy logic for determining optimal drug dosing regimens for treating heart failure, the addition of the safety filter increases the confidence in using this new approach. An example case demonstrating how the fuzzy inference model works is shown by applying it to a previously published clinical case report describing a patient in India who suffered from advanced chronic kidney disease, had experienced multiple episodes of decompensated heart failure and presented with high levels of circulating Nt-Pro BNP. The application of this fuzzy inference system resulted in recommendations to significantly increase the dosage of loop diuretics being administered, but also to begin cautiously administering an Angiotensin Receptor-Nephrilysin Inhibitor at a lower than maximum approved dose due to concerns regarding potential worsening of hyperkalemia and/or hypotension. No recommendation was made to initiate Mineralocorticoid Receptor Antagonist (MRA) therapy. The authors demonstrate that subsequent simulations using follow up data show that if there are improvements in both congestion status and the level of biomarkers (e.g., Nt-Pro BNP), this system may be able to assist clinicians in gradually increasing the dosage of medications while maintaining safety. Ultimately, these results suggest that fuzzy inference systems offer a useful tool for developing clinically meaningful and mathematically flexible approaches to personalize pharmacologic treatment options for individual patients with uncertain conditions.
Keywords: Fuzzy inference system, heart failure management, optimized drug dosing, Mamdani inference, personalized medicine, loop diuretics, ARNI, mineralocorticoid receptor antagonist, clinical decision support, renal safety, NT-proBNP, congestion assessment.
Keywords: Fuzzy inference system, heart failure management, optimized drug dosing, Mamdani inference, personalized medicine, loop diuretics, ARNI, mineralocorticoid receptor antagonist, clinical decision support, renal safety, NT-proBNP, congestion assessment.
Abstract
A Robust Offline-First Architecture for Integrated Examination and Institutional Data Management in Rural Educational Societies
Prof. Diksha Bansod, Shruti Bangare, Vinit Pawankar
DOI: 10.17148/IJARCCE.2026.154139
Abstract: The increasing reliance on digital technologies in education has exposed significant challenges in rural educational environments, particularly due to inconsistent internet connectivity and limited infrastructure. This paper presents a robust offline-first architecture for an integrated examination and institutional data management system designed specifically for rural educational societies. The proposed system enables seamless operation in both offline and online modes by utilizing local data storage and a synchronization mechanism that updates the central database when connectivity is restored.
The system integrates key functionalities such as online examination management, automated evaluation, and institutional information handling within a unified platform. It employs role-based access control to ensure data security and supports efficient data processing even in low-resource settings. Experimental analysis demonstrates that the proposed architecture significantly reduces dependency on continuous internet access while maintaining data consistency and system reliability.
The results indicate that the system improves accessibility, enhances operational efficiency, and supports the digital transformation of rural educational institutions.
Keywords: Offline-First Architecture, Web-Based Examination System, Institutional Information System, Data Synchronization, Rural Education, Local Storage, Role-Based Access Control, Digital Transformation.
The system integrates key functionalities such as online examination management, automated evaluation, and institutional information handling within a unified platform. It employs role-based access control to ensure data security and supports efficient data processing even in low-resource settings. Experimental analysis demonstrates that the proposed architecture significantly reduces dependency on continuous internet access while maintaining data consistency and system reliability.
The results indicate that the system improves accessibility, enhances operational efficiency, and supports the digital transformation of rural educational institutions.
Keywords: Offline-First Architecture, Web-Based Examination System, Institutional Information System, Data Synchronization, Rural Education, Local Storage, Role-Based Access Control, Digital Transformation.
Abstract
AN OVERVIEW ON: AI STOCK FORECASTER: A STOCK PRICE FORECASTING APP
Prof. Priya Farkade, Uday Lanjewar, Ranit Garude, Mohan Khobarkhede, Rohit Bawanukey, Manthan Ukey
DOI: 10.17148/IJARCCE.2026.154140
Abstract: In the past decades, there is an increasing interest in predicting markets among economists, policymakers, academics and market makers. The objective of the proposed work is to study and improve the supervised learning algorithms to predict the stock price. The stock market plays a crucial role in a country's financial growth, but it is also highly dynamic and unpredictable. Stock Market Analysis of stocks using data mining will be useful for new investors to invest in stock market based on the various factors considered by the software. Stock market includes daily activities like Sensex calculation, exchange of shares. The exchange provides an efficient and transparent market for trading in equity, debt instruments and derivatives.
Keywords: Stock Market, ARIMA, LSTM, Linear Regression, Web Interface, Machine Learning, Linear Regression, Python, Django framework, Yahoo Finance.
Keywords: Stock Market, ARIMA, LSTM, Linear Regression, Web Interface, Machine Learning, Linear Regression, Python, Django framework, Yahoo Finance.
Abstract
A STUDY OF PERSONAL FINANCIAL PLANNING AND SAVING BEHAVIOUR OF SALRIED EMPLOYEES 1Shraddha Kadam 2Nikita Khandade 3Janhavi Kasar 4Omkar Khutwad
Dr. Pradyuman Shastri
DOI: 10.17148/IJARCCE.2026.154141
Abstract: This research paper investigates the intricate dynamics of personal financial planning and saving behaviour among salaried employees, a demographic characterized by fixed monthly inflows but increasingly complex financial obligations. In the contemporary economic environment, characterized by inflationary pressures, market volatility, and a shifting reliance from defined-benefit pensions to individual-driven retirement schemes, the necessity for robust financial discipline has never been more critical. The primary objective of this study is to examine how salaried individuals allocate their disposable income, the psychological and socio-economic factors that drive their saving habits, and the extent to which financial literacy influences their long-term wealth creation.
The study adopts a descriptive and analytical research methodology, utilizing primary data collected through a structured questionnaire administered to a diverse sample of salaried professionals across various sectors, including Information Technology, Banking, Healthcare, and Education. To ensure a comprehensive analysis, the research employs several statistical tools, including descriptive statistics, correlation analysis, and regression models, to test hypotheses regarding the relationship between demographic variables—such as age, gender, marital status, and income level—and financial decision-making.
Keywords: Salaried Employees, Saving Behaviour, Financial Planning, Investment Patterns, Financial Literacy.
The study adopts a descriptive and analytical research methodology, utilizing primary data collected through a structured questionnaire administered to a diverse sample of salaried professionals across various sectors, including Information Technology, Banking, Healthcare, and Education. To ensure a comprehensive analysis, the research employs several statistical tools, including descriptive statistics, correlation analysis, and regression models, to test hypotheses regarding the relationship between demographic variables—such as age, gender, marital status, and income level—and financial decision-making.
Keywords: Salaried Employees, Saving Behaviour, Financial Planning, Investment Patterns, Financial Literacy.
Abstract
AN OVERVIEW ON: MEDICAL SYMPTOMS CHECKER APP
Prof. Sonal R Tiwari*, Sagar Kamble, Vaibhav Gawade
DOI: 10.17148/IJARCCE.2026.154142
Abstract: The rapid integration of Artificial Intelligence (AI) and Machine Learning (ML) in healthcare has facilitated the development of automated medical symptom checkers, designed to bridge the gap between initial patient concern and professional clinical diagnosis. This research evaluates the efficacy, diagnostic accuracy, and user-centric design of digital triage platforms that utilize probabilistic modeling and natural language processing (NLP) to interpret patient- reported symptoms. By analyzing large datasets of clinical encounters, these systems aim to provide actionable health insights while mitigating the burden on primary care facilities. However, challenges persist regarding the clinical safety of automated advice and the potential for "cyberchondria" among users. This paper presents a comparative analysis of leading symptom-checker algorithms, highlighting the critical balance between accessibility and diagnostic precision, and concludes with a framework for integrating these tools into the broader telehealth ecosystem to ensure data privacy and evidence-based reliability.
Keywords: Medical Symptom Checker, Artificial Intelligence, Digital Triage, Machine Learning in Medicine, Clinical Decision Support, Telehealth, Diagnostic Accuracy, Health Informatics.
Keywords: Medical Symptom Checker, Artificial Intelligence, Digital Triage, Machine Learning in Medicine, Clinical Decision Support, Telehealth, Diagnostic Accuracy, Health Informatics.
Abstract
AN OVERVIEW ON: AI MEETING INTELLIGENCE A MEETING ANALYZER APP
Prof. Atul Akotkar, Yogesh Puri, Yash Gadling, Shatayu Meshram, Shivdas Lakhe, Piyush Chauhan
DOI: 10.17148/IJARCCE.2026.154143
Abstract: We wish to express our heartfelt gratitude to all the people who have played a crucial role in the research for this project; without their active cooperation, the preparation of this project could not have been completed within the specified time limit We are thankful to our respected Guide, Prof. Atul Akotkar, who supported us throughout this project with utmost cooperation and patience. Thank you for guiding us through the complex integration of AI models and for motivating us to complete this project with complete focus and attention. We are also thankful to the Head of Department, Prof. Atul Akotkar, and Principal, Prof.Rasik Upadhye. It is their kind help and support that have made our study and life a wonderful time. Finally,
Abstract
AI - POWERED HUMAN RESOURCE MANAGEMENT SYSTEM (HRMS)
Bina Rewatka, Rohit Umredkar, Vaishnavi Tandekar, Priyanka Choudhari, Shrishakti Sharma
DOI: 10.17148/IJARCCE.2026.154144
Abstract: The Human Resource Management System (HRMS) is a full-stack web-based application designed to automate and streamline core HR operations such as employee management, attendance tracking, payroll processing, recruitment, and performance evaluation. Traditional HR systems often rely on manual processes, which are time- consuming, error-prone, and lack real-time decision-making capabilities.
This project introduces an AI-powered HRMS integrated with real-time analytics to enhance efficiency and decision- making in human resource operations. The system provides a centralized platform where HR administrators can manage employee records, track attendance, process payroll, and monitor performance metrics seamlessly.
An intelligent chatbot is integrated to assist users with HR-related queries such as leave balance, company policies, and employee information, improving accessibility and user experience. Real-time updates are enabled using modern web technologies, ensuring instant synchronization of data across the system.
The application is built using a modern full-stack architecture, with React.js for the frontend, Node.js and Express.js for backend services, and MongoDB/PostgreSQL for database management. The system is deployed on cloud infrastructure using Docker and CI/CD pipelines for scalability and reliability.
Keywords: AI-powered HRMS, Human Resource Management System, Real-time analytics, Employee management, Payroll system, Attendance tracking, Recruitment system, Chatbot, Web application.
This project introduces an AI-powered HRMS integrated with real-time analytics to enhance efficiency and decision- making in human resource operations. The system provides a centralized platform where HR administrators can manage employee records, track attendance, process payroll, and monitor performance metrics seamlessly.
An intelligent chatbot is integrated to assist users with HR-related queries such as leave balance, company policies, and employee information, improving accessibility and user experience. Real-time updates are enabled using modern web technologies, ensuring instant synchronization of data across the system.
The application is built using a modern full-stack architecture, with React.js for the frontend, Node.js and Express.js for backend services, and MongoDB/PostgreSQL for database management. The system is deployed on cloud infrastructure using Docker and CI/CD pipelines for scalability and reliability.
Keywords: AI-powered HRMS, Human Resource Management System, Real-time analytics, Employee management, Payroll system, Attendance tracking, Recruitment system, Chatbot, Web application.
Abstract
Conflict-Free Replicated Data Types: An Exhaustive Analysis of Theoretical Foundations, Synchronization Protocols, and State-of-the-Art Architectures
Vivek R, Panchami M Hegde
DOI: 10.17148/IJARCCE.2026.154145
Abstract: The architectural topography of contemporary distributed computing is strictly governed by the intricate mathematical balance between data availability, partition tolerance, and stringent consistency. As computational systems increasingly expand into geographically distributed cloud platforms, edge-deployed collaborative networks, and high- frequency real-time databases, the mandate for fault tolerance and ultra-low latency access dictates that data must be asynchronously replicated across multiple network nodes. However, adhering to the fundamental constraints of the CAP theorem, distributed database architects have historically deferred to centralized consensus-based algorithms, such as Paxos and Raft, which ensure strong consistency through replicated state machines but inherently sacrifice availability during inevitable network partitions. To circumvent these prohibitive latency bottlenecks, the paradigm of Optimistic Replication emerged, eventually crystalizing into the mathematically rigorous framework of Conflict-Free Replicated Data Types (CRDTs). By formalizing a Strong Eventual Consistency (SEC) model, CRDTs guarantee that any two replicas receiving the identical set of updates will deterministically converge to an exact, unified state without ever requiring global coordination. This exhaustive research report dissects the comprehensive landscape of CRDT architectures, examining the foundational algebraic literature formalized across seminal academic research spanning nearly two decades. It provides a highly detailed, authoritative analysis of state-based (CvRDT) and operation-based (CmRDT) synchronization frameworks, causality tracking mechanisms utilizing logical and probabilistic Bloom clocks, and the chronological algorithmic evolution of sequence data types optimized for decentralized collaborative text editing. Furthermore, the analysis explores modern advancements in metadata compression via Delta-CRDTs, the integration of Byzantine Fault Tolerance utilizing cryptographic hash graphs and universal Blocklace structures, and the state-of-the- art expansion of CRDT logic into complex geometric topologies for geospatial mapping systems and structured relational databases preserving strict SQL invariants.
Keywords: Conflict-Free Replicated Data Types, Strong Eventual Consistency, Optimistic Replication, Distributed Systems, Byzantine Fault Tolerance, Collaborative Editing, Vector Clocks, Join-Semilattices, Delta-CRDTs, Eventual Consistency.
Keywords: Conflict-Free Replicated Data Types, Strong Eventual Consistency, Optimistic Replication, Distributed Systems, Byzantine Fault Tolerance, Collaborative Editing, Vector Clocks, Join-Semilattices, Delta-CRDTs, Eventual Consistency.
Abstract
Effect of Image Pre-Processing Techniques on Object Detection
Aparna M, Dr. H Mary Shyni
DOI: 10.17148/IJARCCE.2026.154146
Abstract: This work analyses how different image pre-processing techniques affect object detection results. Objects are detected after applying various processing methods, and the outputs are compared. Most existing studies focus on improving object detection models, but there is limited work analysing how individual pre-processing techniques influence detection results. YOLOv8 (You Only Look Once) is used in this study as it is a fast and reliable model pre- trained on the COCO (Common Objects in Context) dataset. The focus of this work is to observe how different techniques affect the number of detected objects and their corresponding confidence scores, which indicate how certain the model is about a detected object and its location. The effects of techniques such as greyscale conversion, histogram equalisation, contrast adjustment, blurring, and edge detection are analysed on real-world images and a subset of the COCO dataset. Approximately 20 images from the coco128 subset are used for controlled analysis. The evaluation is based on a relative comparison of detection outputs using detection count and confidence scores (above a threshold of 0.5). It is observed that different techniques perform differently depending on the image, and no single method consistently provides the best results. This study helps in understanding how pre-processing influences object detection behaviour and supports better selection of techniques based on the input image.
Keywords: Object Detection, YOLOv8, COCO Dataset, Image Pre-processing, Confidence Scores, Histogram Equalisation, Edge Detection
Keywords: Object Detection, YOLOv8, COCO Dataset, Image Pre-processing, Confidence Scores, Histogram Equalisation, Edge Detection
Abstract
BIKE ON RENT: A WEB-BASED BIKE RENTAL MANAGEMENT SYSTEM
Atul Akotkar, Reena Kakde, Himanshu Nagpure, Nikita Dhote
DOI: 10.17148/IJARCCE.2026.154147
Abstract: The Bike On Rent system is a full-stack web-based application designed to simplify and automate the process of renting bikes. Traditional bike rental systems are often manual, time-consuming, and lack real-time availability tracking, which leads to inefficiencies and poor user experience.
This project introduces a modern bike rental platform developed using the MERN stack (MongoDB, Express.js, React.js, Node.js), providing a centralized system for users, vendors, and administrators. The system allows users to browse available bikes, check details, and book rentals online with ease. Vendors can manage bike inventory, update availability, and track bookings through a dedicated dashboard.
Real-time updates are enabled to ensure accurate bike availability and seamless booking experience. Secure authentication mechanisms are implemented to protect user data and transactions. The system also includes admin control for monitoring and managing overall platform activities.
Overall, the application improves efficiency, reduces manual effort, and provides a scalable and user-friendly solution for bike rental services.
Keywords: Bike Rental System, MERN Stack, Web Application, Online Booking, Rental Management
This project introduces a modern bike rental platform developed using the MERN stack (MongoDB, Express.js, React.js, Node.js), providing a centralized system for users, vendors, and administrators. The system allows users to browse available bikes, check details, and book rentals online with ease. Vendors can manage bike inventory, update availability, and track bookings through a dedicated dashboard.
Real-time updates are enabled to ensure accurate bike availability and seamless booking experience. Secure authentication mechanisms are implemented to protect user data and transactions. The system also includes admin control for monitoring and managing overall platform activities.
Overall, the application improves efficiency, reduces manual effort, and provides a scalable and user-friendly solution for bike rental services.
Keywords: Bike Rental System, MERN Stack, Web Application, Online Booking, Rental Management
Abstract
Smart Traffic Management Web System Using AI and Real-Time Analytics
Varsha Santosh Ekhande
DOI: 10.17148/IJARCCE.2026.154148
Abstract: Urban traffic congestion has become a major challenge in rapidly growing cities, leading to increased travel time, fuel consumption, and environmental pollution. Traditional traffic control systems are mostly static and do not adapt to real-time conditions, which results in inefficient traffic flow. This paper presents a Smart Traffic Management Web System that integrates real-time monitoring, artificial intelligence, and adaptive signal control to improve overall traffic efficiency.
The proposed system uses computer vision techniques based on YOLO and OpenCV to detect vehicles from video input and estimate traffic density. A dynamic signal control mechanism adjusts traffic light timing based on congestion levels. The system also provides a web-based dashboard for monitoring traffic conditions, analytics visualization, and emergency vehicle prioritization. Firebase is used as the backend for real-time data storage and synchronization.
The proposed system uses computer vision techniques based on YOLO and OpenCV to detect vehicles from video input and estimate traffic density. A dynamic signal control mechanism adjusts traffic light timing based on congestion levels. The system also provides a web-based dashboard for monitoring traffic conditions, analytics visualization, and emergency vehicle prioritization. Firebase is used as the backend for real-time data storage and synchronization.
Abstract
A Severity-Aware Hybrid ML Model for Real- Time Cyber Threat Detection and Alerting
Bhavani Kothapalli, Bodapati Preethi, Brahma K, K P S Kavya
DOI: 10.17148/IJARCCE.2026.154149
Abstract: The increasing dependence on digital authentication systems has increased the risk of unauthorized access and abnormal behavior. Many traditional security systems use fixed rules, which are not effective in detecting new or hidden cyber attacks. This paper presents a severity-aware hybrid machine learning system for real-time cyber threat detection and alerting based on login behavior. The system uses a Random Forest model along with an Isolation Forest model to identify suspicious login activities. Important features such as login time, location, failed attempts, and behavioral patterns are analyzed. Based on the level of risk, the system classifies threats into Low, Medium, and High severity levels. All detected threats are stored and displayed using an interactive dashboard, helping administrators monitor and respond to security issues effectively.
Keywords: Cybersecurity, Login Behavior Analysis, Anomaly Detection, Machine Learning, Random Forest, Isolation Forest, Severity Classification
Keywords: Cybersecurity, Login Behavior Analysis, Anomaly Detection, Machine Learning, Random Forest, Isolation Forest, Severity Classification
Abstract
AI-Based Smart Healthcare Assistance System
Prof. Diksha Bansod, Sneha K. Shrirame, Triveni M. Kirsan, Pranav S. Machave, Payal A., Uikey, Aditya A. Langade
DOI: 10.17148/IJARCCE.2026.154150
Abstract: The healthcare landscape in India faces a unique set of challenges: a large patient-to-doctor ratio, limited physical access to specialists in rural areas, and high out-of-pocket costs. Digital tools that can partially address these gaps have become increasingly relevant. This paper presents an AI-Based Smart Healthcare Assistance System—a web- based platform built on the Django framework that integrates three conversational AI modules and a doctor appointment booking subsystem into a single application. The core modules include a symptom-based disease predictor using a trained random forest classifier with cosine-similarity-based input matching, a personalised Indian diet plan generator powered by a large language model (GPT-3.5-turbo), and a general health Q&A interface. Building on these, we have added a doctor appointment booking module where doctors and patients can register separately, doctors can manage their schedules and appointment requests from a dedicated dashboard, and patients can search for available doctors by specialisation, send appointment requests, track status, and cancel if needed. The system was tested for functional correctness, classification accuracy, and basic usability. Results show that the random forest classifier achieves 100% accuracy on the test set and all appointment lifecycle states function as expected. The platform demonstrates how a comparatively simple web stack can deliver meaningful AI-assisted healthcare access.
Keywords: Healthcare chatbot, disease prediction, random forest, diet planning, doctor appointment, Django, NLP, machine learning, telemedicine, GPT.
Keywords: Healthcare chatbot, disease prediction, random forest, diet planning, doctor appointment, Django, NLP, machine learning, telemedicine, GPT.
Abstract
Multimodal Deepfake Detection Using Deep Learning
Dhanushree B V, Panchami M Hegde
DOI: 10.17148/IJARCCE.2026.154151
Abstract: Deepfake technology has evolved rapidly, enabling the creation of highly realistic manipulated images, audio, and videos. While these advancements have applications in entertainment and media, they also pose significant risks such as misinformation, identity fraud, and security threats. This research focuses on multimodal deepfake detection using deep learning techniques by combining visual and audio features for improved accuracy. The proposed approach integrates Convolutional Neural Networks (CNNs) for image analysis and Natural Language Processing (NLP) and audio-based models for detecting inconsistencies across modalities. By leveraging multimodal data, the system enhances detection robustness compared to unimodal approaches. Experimental results demonstrate that combining visual and audio cues significantly improves detection performance and generalization across different types of deepfakes. This system can be applied in social media monitoring, digital forensics, and cybersecurity applications.
Keywords: Deepfake Detection, Multimodal Learning, Machine Learning, CNN, NLP, Audio-Visual Analysis, Artificial Intelligence
Keywords: Deepfake Detection, Multimodal Learning, Machine Learning, CNN, NLP, Audio-Visual Analysis, Artificial Intelligence
Abstract
Virtual Reality–Integrated Telepresence Robot for Intelligent Remote Monitoring
Dr M V Sreenivas Rao, Jia M S, Preethu K
DOI: 10.17148/IJARCCE.2026.154152
Abstract: In the current day and age, telepresence has gained more and more recognition, and individuals cannot be physically present at two or more places simultaneously. Telepresence systems are already being used in surveillance and dangerous places where the lives of individuals are at risk. This project aims to build a gesture-controlled virtual telepresence robot, which can be used in defense, mining, and educational scenarios. The system will include a robot with a camera for real-time video streaming, a VR headset with a smartphone for viewing the video, and gesture-controlled gloves for controlling the robot’s movement. A Raspberry Pi will be used for real-time video streaming over Wi-Fi to the smartphone, which will be inserted into the VR headset. The robot will be controlled with the help of hand gestures, which will be detected with the help of an MPU6050 accelerometer connected to the user’s glove. The gestures will be detected with the help of Arduino Nano over Bluetooth and processed with the help of an Arduino Uno, which will control the robot with the help of a motor driver. A servo motor will be used to rotate the camera up to 180 degrees.
Keywords: virtual telepresence, remote controlled, VR, hand gloves, surveillance.
Keywords: virtual telepresence, remote controlled, VR, hand gloves, surveillance.
Abstract
CyberWatch: A Real-Time Intelligent Web Security Monitoring and Attack Detection Platform
Drupad Dhamdhere, Shardul Pharande, Aashish Kondgire, Sakshi Phutane, Prof. Supriya Jagtap
DOI: 10.17148/IJARCCE.2026.154153
Abstract: CyberWatch is a real-time web security monitoring platform built to give defenders immediate visibility into web attacks. Implemented on Spring Boot, it inspects incoming requests and applies modular detectors for SQL injection, brute-force abuse, XSS, command injection, DDoS, CSRF, SSRF, malicious file upload, XXE, and LDAP injection. The implementation also includes log-injection and directory-traversal checks for broader operational coverage. All primary attack categories in project scope were implemented and validated using controlled test traffic. The platform captures contextual evidence, computes a weighted risk score, and streams prioritized alerts to a WebSocket dashboard for faster triage. Experimental results show a strong precision-recall balance with low latency overhead, indicating practical suitability for academic environments and small-to-mid production deployments.
Keywords: Web application security, attack detection, Spring Boot, real-time dashboard, cyber threat monitoring.
Keywords: Web application security, attack detection, Spring Boot, real-time dashboard, cyber threat monitoring.
Abstract
POTHOLE & CRACK DETECTION SYSTEM USING ML & COMPUTER VISION
Prof. Amit Meshram, Komal Rewaskar, Pratiksha Tidke, Tannu Rangarkar, Akshada Sable, Tanushree Dhote
DOI: 10.17148/IJARCCE.2026.154154
Abstract: This project aims to develop an automated system that can detect potholes and cracks on roads using image processing and machine learning techniques. Roads are an important part of daily transportation, but damages like potholes and cracks can cause accidents and affect vehicle movement. Traditionally, road inspection is done manually, which takes a lot of time and effort. To solve this problem, this project presents a system that can automatically detect potholes and cracks using machine learning and computer vision.
The system uses images of roads as input and processes them using deep learning techniques such as Convolutional Neural Networks (CNN) and object detection models like YOLO. The model is trained on a dataset of road images containing different types of damages. Once trained, it can identify and highlight potholes and cracks in both images and real time video.
The proposed system helps in faster and more accurate detection compared to manual methods. It can be Useful for road maintenance authorities to monitor road condition and take timely action. In the future, this system can be improved by adding GPS tracking and mobile based application for real time reporting.
Keywords: Potholes, Cracks, Road Damage, Machine Learning, Computer Vision, Image Processing, Road Safety.
The system uses images of roads as input and processes them using deep learning techniques such as Convolutional Neural Networks (CNN) and object detection models like YOLO. The model is trained on a dataset of road images containing different types of damages. Once trained, it can identify and highlight potholes and cracks in both images and real time video.
The proposed system helps in faster and more accurate detection compared to manual methods. It can be Useful for road maintenance authorities to monitor road condition and take timely action. In the future, this system can be improved by adding GPS tracking and mobile based application for real time reporting.
Keywords: Potholes, Cracks, Road Damage, Machine Learning, Computer Vision, Image Processing, Road Safety.
Abstract
Development of a Piezoelectric Footstep Power Generation System with Arduino-Based Monitoring
Karthi Balaji M, Suraj K, Rudrakash Agarwal, Nithila K, Sandhya Dillybabu, Ms Charulatha R.T
DOI: 10.17148/IJARCCE.2026.154155
Abstract: Recently, the demand for sustainable, renewable energy resulted in exploration of alternate energy-harvesting approaches. This paper describes a design and construction approach to a low-cost Footstep Power Generation system that uses piezoelectric transducers that helps to convert the mechanical energy generated by human footsteps into electrical energy. Six 35mm piezoelectric discs are arranged under a walking surface and connected through a rectifier circuit which comprises 1N4007 diodes, a 10μF capacitor, and a BC547 transistor to stipulate the harvested AC signal into a stable DC output. An Arduino Uno microcontroller reads the output voltage via analog pin A0, counts footsteps using edge-detection logic, and displays both parameters in real time on a 16×2 I2C LCD display. A 2×18650 lithium- cell battery pack serves as supplementary power and an optional storage medium. Experimental results verify that the prototype can successfully harvest energy, count steps, and measure voltage, with the LCD showing the number of steps and voltage in millivolts for each footstep event.The system is adequate to prove that piezoelectric transduction is one viable means of harvesting energy from areas where there is a high footfall, such as corridors, stairways, or public walkways.
Keywords: Piezoelectric Transducer; Energy Harvesting; Arduino Uno; Footstep Power; IoT Monitoring; Renewable Energy; LCD Display; BC547; Rectifier Circuit.
Keywords: Piezoelectric Transducer; Energy Harvesting; Arduino Uno; Footstep Power; IoT Monitoring; Renewable Energy; LCD Display; BC547; Rectifier Circuit.
Abstract
IOT PROJECT: PEOPLE COUNTER USING IR SENSORS
SREEJITH S, DURAI RAJ R, MANISH KUMAR MANDAL, ARAVIND SRIRAM, DEEPAK G, MS . CHARULATHA R T
DOI: 10.17148/IJARCCE.2026.154156
Abstract: A people counting system was designed and implemented using infrared (IR) sensors to accurately track the number of individuals entering and exiting an enclosed space in real time. The system was simulated and tested using Cisco Packet Tracer version 8.0.0 alongside embedded hardware prototyping tools. Two IR sensors are placed at the entry and exit points of a room; when a person passes through, the sensors detect interruptions in the infrared beam and increment or decrement a counter accordingly. The count data is processed by a microcontroller and transmitted over the local network to a centralized server for monitoring. The server is accessible only through a secured HTTPS connection on port 443. Internally, department routers are interconnected for seamless access to the monitoring application. Network Address Translation (NAT) is configured on the company router to map public IP addresses to private device addresses. Switches are used to optimize device connectivity across both the campus network and the broadband connection. Connectivity was verified using ping tests from all segments of the network to confirm full reachability of the people counter server and monitoring dashboard
Abstract
Flood AI Monitoring & Early Warning System: A Machine Learning and IoT Integrated Approach Using CWC Data Sandipkumar C. Sagare¹, Ahilya Hapate², Prachi Chougule³, Pranoti Athawale⁴, Rakhi
Adgane, Sayka Arab
DOI: 10.17148/IJARCCE.2026.154157
Abstract: Floods rank among the world’s most catastrophic natural disasters, inflicting severe loss of life, destruction of infrastructure, and socioeconomic disruption. Conventional flood monitoring systems rely on manual observation and static threshold-based alerts, which lack real-time intelligence and fail to provide adequate lead time for evacuation or mitigation. This paper presents a comprehensive AI-powered flood prediction and early warning system that integrates machine learning with IoT-based real-time monitoring. Using hourly river water-level telemetry from the Central Water Commission (CWC) of India, a Random Forest Classifier is trained on engineered temporal features including rate of rise, rolling averages, and danger-level proximity. An ESP32 microcontroller paired with an ultrasonic sensor provides live field measurements, which are processed by a Flask/FastAPI backend and visualized on an interactive dashboard with Twilio-based SMS/email alerts. The system achieves 96% overall accuracy and 97% flood-class precision. Critical analysis of the class-imbalance challenge is provided, with a roadmap for improving recall through SMOTE oversampling and deep learning architectures.
Keywords: Flood prediction, Random Forest, IoT, ESP32, CWC, class imbalance, early warning system, machine learning, SMOTE, real-time monitoring
Keywords: Flood prediction, Random Forest, IoT, ESP32, CWC, class imbalance, early warning system, machine learning, SMOTE, real-time monitoring
Abstract
INDUSTRIAL FAULT MONITORING SYSTEM
Mr. M. RAVI, M. Tech(Ph. D), T. VENU MADHAV*
DOI: 10.17148/IJARCCE.2026.154158
Abstract: Industrial operations often face challenges due to unexpected machine failures, inefficient monitoring, and lack of real-time fault detection, which can lead to production loss, increased maintenance costs, and safety risks. Traditional fault monitoring methods rely on manual inspection and periodic maintenance, making them ineffective for early fault detection and continuous supervision. Therefore, there is a growing need for an intelligent system that can automate fault monitoring and enhance industrial efficiency.
This paper presents the design and implementation of an Industrial Fault Monitoring System using IoT and embedded intelligence. The proposed system integrates various sensors such as temperature sensors for detecting overheating, vibration sensors for identifying mechanical faults, and current/voltage sensors for monitoring electrical conditions. An embedded platform such as Raspberry Pi or microcontroller is used to process real-time sensor data.
Machine learning techniques are employed to analyze machine behavior, detect anomalies, and predict potential faults based on historical and real-time data. The system continuously monitors machine parameters and automatically updates the system status whenever abnormal conditions are detected. Users can remotely access machine data through a mobile application enabled by IoT connectivity.
In addition, the system provides instant alerts for fault conditions such as overheating, excessive vibration, or electrical overload, ensuring timely corrective actions. It can also assist in predictive maintenance by analyzing performance trends and suggesting maintenance schedules.
The proposed Industrial Fault Monitoring System offers a reliable, efficient, and cost-effective solution for modern industries. By combining IoT, sensor integration, and intelligent data analytics, it enhances machine reliability, improves safety, and supports smart industrial automation in Industry 4.0 environments.
Keywords: Industrial Fault Monitoring System, Internet of Things (IoT), Machine Learning, Embedded Systems, Raspberry Pi, ESP32, Temperature Sensor, Vibration Sensor, Current Sensor, Real-Time Monitoring, Predictive Maintenance, Wireless Communication, Cloud Computing, Industrial Automation, Smart Factory, Fault Detection, Remote Monitoring, Data Analytics, Industry 4.0, Safety Monitoring.
Abstract
IoTCaloTrack: An IoT-Enabled Real-Time Calorie Tracking System
Akshat Kumar, Sauban Tarique, Preetam N, Yogesh P, Tharun VS, Ms. Charulatha R.T
DOI: 10.17148/IJARCCE.2026.154159
Abstract: In today’s health-conscious world and simultaneously the world of unhealthy and processed foods, managing what a person eats, how much a person eats and most of all how much it affects their body is getting harder as time goes on. In this paper, IoTCaloTrack, a novel Internet-of-Things framework for real-time calorie monitoring using a web frontend and Python FastAPI backend is introduced. This system integrates a microcontroller with a camera and load- cell sensor, a FastAPI server, and the Google Gemini AI for food recognition. In a typical workflow, a user triggers meal capture via the web interface; the IoT device then captures a photo of the food and measures its weight. The backend then uploads the image to the Gemini API, which identifies the dish. Knowing the food type and its measured weight, IoTCaloTrack computes caloric content by multiplying weight (in grams) by a nutrient database value. The result is stored and returned to the web client. Gemini’s high accuracy (20% improved recognition over prior methods) ensures reliable identification. Experiments show that Gemini 2.5 processes images very quickly (≈1 second faster than earlier AI models), enabling near-instant calorie estimates. It is a comprehensive calorie management and tracking system where each user can independently store their nutritional data and access it conveniently.
Keywords: IoT, FastAPI, Web Frontend, Calorie Tracking, Food Recognition, Gemini API, Python.
Keywords: IoT, FastAPI, Web Frontend, Calorie Tracking, Food Recognition, Gemini API, Python.
Abstract
Audio Video Dubbing System
Saakshi Nalawade, Shruti Namaye, Sahil Raut, Shreyash Sawant, Manoj M. Deshpande
DOI: 10.17148/IJARCCE.2026.154160
Abstract: With the rapid growth of digital media, the demand for multilingual audio video content has increased significantly. Traditional dubbing techniques are time consuming, costly, and require extensive human effort. This paper presents an AI based audio video dubbing system that automatically converts spoken content from one language to another while preserving the original speaker voice characteristics and synchronizing with the video. The proposed system integrates speech to text conver- sion, neural machine translation, and text to speech synthesis to generate natural and realistic dubbed audio. Voice cloning techniques are used to maintain speaker identity across different languages. Additionally, audio alignment is applied to ensure smooth synchronization between the generated speech and the video stream. The system aims to provide an efficient and scalable solution for content localization in applications such as education, entertainment, and online media platforms. Experimental results demonstrate that the proposed approach significantly reduces manual effort while maintaining acceptable audio quality and intelligibility.
Keywords: Audio video dubbing, voice cloning, speech synthesis, machine learning, multilingual translation, lip synchro- nization
Keywords: Audio video dubbing, voice cloning, speech synthesis, machine learning, multilingual translation, lip synchro- nization
Abstract
A Deterministic Multi-Metric Framework for Automated Image Dataset Validation in Computer Vision
M.Balavignesh, Dr. C. Karpagavalli, Dr. M. Kaliappan
DOI: 10.17148/IJARCCE.2026.154161
Abstract: Deep learning model performance in computer vision is fundamentally limited by the quality of training data, yet augmented datasets frequently contain feature corruption such as extreme blur, noise, and lighting anomalies. This paper presents the AI-Based Image Dataset Quality Validator, a high-precision, data-centric framework designed for automated dataset sanitization. The system employs a deterministic multi-metric validation pipeline integrating Laplacian Variance for sharpness auditing and ITU-R 601 Luma weighting for exposure control, enabling fine-grained defect identification that traditional global-threshold filters miss. A core innovation of the architecture is the Parallel Structural Label Synchronization module, which guarantees a strict 1:1 correspondence between images and their respective annotations stored in either TXT or CSV format, automatically eliminating orphan labels during export. To handle large- scale batches on standard hardware, the system implements Active Memory Recovery through controlled garbage collection. Experimental evaluation on a 500-image benchmark demonstrates 96.8% rejection accuracy with an average throughput of 42.5 ms per image. The proposed framework reduces manual data-cleaning effort by an estimated 98%, delivering a scalable, Green AI solution for high-integrity computer vision pipelines.
Keywords: Image Dataset Validation, Computer Vision, Laplacian Variance, Label Synchronization, Data-Centric AI, Image Quality Assessment (IQA), YOLO Framework, Green AI, Automated Data Sanitization, TXT/CSV Annotation Management.
Keywords: Image Dataset Validation, Computer Vision, Laplacian Variance, Label Synchronization, Data-Centric AI, Image Quality Assessment (IQA), YOLO Framework, Green AI, Automated Data Sanitization, TXT/CSV Annotation Management.
Abstract
Comprehensive Analysis of Proactive Fraud Prevention and Contribution of Machine Language
Harshit Sharma, Himanshu Chauhan, Muskan Sharma, Mayank Sharma, Dr. Satish Kumar Soni, Dr. Uruj Jaleel
DOI: 10.17148/IJARCCE.2026.154162
Abstract: The rising trend of fraud is considered to be one of the most significant issues in the contemporary digital environment, especially in finance, e-commerce, insurance and telecom sectors. Conventional fraud detection approaches are primarily based on rule-based frameworks, whose effectiveness deteriorates as fraud evolves. In contrast to traditional approaches, machine learning (ML) and predictive analytics offer flexible and data-centric ways to prevent fraud proactively. This study focuses on how ML techniques such as supervised, unsupervised and deep learning can be utilized to detect.
Abstract
A STUDY ON MATERIAL DISPATCH AND DOCUMENTATION PROCESSES IN A MANUFACTURING COMPANY
Anish Nivas Shidruk, Dr. Pradyuman Shastri
DOI: 10.17148/IJARCCE.2026.154163
Abstract: This research paper examines the material dispatch and documentation processes in manufacturing companies, which are critical components of supply chain and logistics management. In today’s competitive industrial environment, efficient handling of dispatch operations ensures timely delivery, cost optimization, and customer satisfaction. The study focuses on understanding the procedures involved in material dispatch, the documentation required, and the challenges faced in maintaining accuracy and efficiency.
The research adopts a descriptive and analytical methodology using primary data collected through structured questionnaires and interviews with employees working in logistics, warehouse, and dispatch departments. Secondary data is gathered from company records, logistics reports, and industrial publications. Statistical tools are used to analyze the efficiency, accuracy, and reliability of dispatch operations.
The study highlights the importance of proper documentation such as invoices, delivery challans, transport receipts, and compliance documents in ensuring smooth operations and minimizing errors.
Keywords: Material Dispatch, Documentation Process, Manufacturing Industry, Logistics Management, Supply Chain.
The research adopts a descriptive and analytical methodology using primary data collected through structured questionnaires and interviews with employees working in logistics, warehouse, and dispatch departments. Secondary data is gathered from company records, logistics reports, and industrial publications. Statistical tools are used to analyze the efficiency, accuracy, and reliability of dispatch operations.
The study highlights the importance of proper documentation such as invoices, delivery challans, transport receipts, and compliance documents in ensuring smooth operations and minimizing errors.
Keywords: Material Dispatch, Documentation Process, Manufacturing Industry, Logistics Management, Supply Chain.
Abstract
POINT OF USE & POINT OF ENTRY IN WATER TREATMENT
Prof. Nikhil Gahukar, Prof. Vrushali Paraye, Prof. Sameena Ansari, Prof. Mohammad Rafiullah, Prof., Swati Jadhav, Honey A. Bakshi, Dipti V. Bawane, Vaibhav L. Dhurve.
DOI: 10.17148/IJARCCE.2026.154164
Abstract: This project presents the design and development of an E-commerce platform Provision of safe and reliable drinking water remains one of the most critical challenges for public health, particularly in small communities, rural settlements, and decentralized habitations where installation of large centralized treatment facilities is economically and technically difficult. Point- of-Use (POU) and Point-of-Entry (POE) water treatment systems have emerged as practical, cost- effective, and adaptable alternatives to conventional treatment infrastructure. These systems treat water either at the point where it enters a household (POE) or at the point where it is consumed, such as a tap or kitchen outlet (POU). The present synopsis examines the technological principles, regulatory background, performance capabilities, and implementation strategies associated with POU/POE devices. Various treatment technologies including activated carbon filtration, reverse osmosis, ion exchange, distillation, ultraviolet disinfection, and aeration are evaluated with respect to their contaminant removal efficiency, operational requirements, maintenance needs, and economic feasibility. Special attention is given to their ability to meet drinking water quality standards prescribed under national and international guidelines. The study also highlights institutional responsibilities, monitoring challenges, and long-term sustainability issues, as proper operation and maintenance determine the success of decentralized treatment approaches. While POU/POE systems offer significant advantages such as lower capital investment, rapid deployment, and targeted contaminant removal, they also present limitationAs related to compliance assurance, user awareness, and waste disposal. This work aims to provide a comprehensive technical foundation for planners, engineers, and decision makers considering POU/POE technologies as viable solutions for achieving safe drinking water delivery. The outcome supports informed selection, improved management practices, and enhanced public health protection
Keywords: POU, POE, Water Treatment, Reverse Osmosis, Filtration, Drinking Water Safety.
Keywords: POU, POE, Water Treatment, Reverse Osmosis, Filtration, Drinking Water Safety.
Abstract
AI-Driven Gesture Tracking: A Comprehensive Review of Techniques, Applications, and Future Directions
Huda Khan, Isha Kaushal, Gunjan Rani, Nikhil Kumar, Mr. K.S. Mishra
DOI: 10.17148/IJARCCE.2026.154165
Abstract: Gesture tracking, the technological ability to interpret human movements as commands, has become a cornerstone of modern Human-Computer Interaction (HCI). Propelled by advances in artificial intelligence (AI), particularly deep learning, gesture recognition systems have evolved from laboratory experiments to practical applications in robotics, manufacturing, healthcare, and consumer electronics. This paper provides a comprehensive review of AI techniques for gesture tracking, spanning the last decade of research. We systematically analyze the gesture recognition pipeline, from data acquisition methods (vision-based, sensor-based) to feature extraction and classification algorithms. The review contrasts traditional machine learning approaches like Support Vector Machines (SVMs) and Hidden Markov Models (HMMs) with modern deep learning architectures, including Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Transformers. A significant focus is placed on training strategies, such as multi-modal fusion and ModDrop, which enhance robustness in real-world conditions. Furthermore, we explore key application domains—Human-Robot Interaction (HRI), Industry 5.0, Augmented Reality (AR), and drone control— highlighting how AI techniques are tailored to meet specific domain challenges. The review concludes by identifying persistent challenges, including occlusion, environmental variability, and the need for large, annotated datasets, and proposes future research directions towards more adaptive, multi-modal, and human-centric gesture recognition systems.
Keywords: Gesture Recognition, Human-Computer Interaction, Deep Learning, Computer Vision, Human-Robot Interaction, Industry 5.0, Multi-modal Fusion
Keywords: Gesture Recognition, Human-Computer Interaction, Deep Learning, Computer Vision, Human-Robot Interaction, Industry 5.0, Multi-modal Fusion
Abstract
EDUCONNECT PORTAL: A WEB-BASED COMMUNICATION SYSTEM FOR PARENTS, TEACHERS AND ADMINISTRATORS
P.P Garate, S.K Gavli, V.A Ghag, V.S Gaikwad, V.A Bhamre
DOI: 10.17148/IJARCCE.2026.154166
Abstract: EduConnect Portal is a web-based academic communication platform developed to strengthen interaction among parents, teachers, and administrators. The system focuses on centralized communication, timely announcements, stakeholder coordination, and accessible digital interaction through a unified web interface. The attached website presents the concept of a real-time portal that simplifies the exchange of educational information and supports transparent academic engagement. This paper describes the motivation, objectives, architecture overview, functional modules, design considerations, benefits, and future scope of the portal. The project demonstrates how front-end web technologies can be applied to solve common communication problems in schools and colleges while creating a scalable base for broader education management services.
Keywords: EduConnect Portal, Parent-Teacher Communication, Educational Website, Web Application, School Management System.
Keywords: EduConnect Portal, Parent-Teacher Communication, Educational Website, Web Application, School Management System.
Abstract
Evolutionary Deep Learning Approach for Driver Drowsiness Detection Using CNN and Genetic Algorithm
Bellamkonda Venkata Sudheer kumar, Dr. Shaik Javed Parvez
DOI: 10.17148/IJARCCE.2026.154167
Abstract: Deep learning has advanced significantly with the Convolutional Neural Network (CNN) and Genetic Algorithm (GA) designs, especially in image recognition and sequential data processing. Biometric data like heart rate, pulse waves, brain waves, and eye movements are the mainstays of traditional drowsiness detection techniques. The technology may identify tiny indicators of exhaustion, such as variations in eyelid movement, eye closure rates, and facial expressions, by examining real-time visual data from a driver's face and eyes. Furthermore, the technology ensures prompt intervention and improves driver safety by providing real-time voice alarms when it detects indicators of drowsiness. Thus, the combination of CNN and GA provides a very effective, scalable, and real-time way to avoid driving accidents caused by drowsiness.
Keywords: Convolutional Neural Network (CNN) , Genetic Algorithm (GA), Deep Learning, Real-Time MonitoringI .
Keywords: Convolutional Neural Network (CNN) , Genetic Algorithm (GA), Deep Learning, Real-Time MonitoringI .
Abstract
AROS: An Intelligent Arctic Route Optimization System Integrating Machine Learning with Environmental Risk Evaluation
Tejas Mali, Abhay Kulkarni, Varun Rasal, Ritik Gohil, Smita Chunamari
DOI: 10.17148/IJARCCE.2026.154168
Abstract: Abstract—Shrinking Arctic ice has opened corridors that were impassable until recently, and operators are paying atten-tion. The Northern Sea Route cuts voyage distances between northern Europe and northeast Asia by 30 to 40 percent compared to the Suez Canal path — a saving hard to ignore. But sailing these waters is a different challenge altogether. Ice concentrations can shift from clear to dangerous in a matter of hours, icebergs calve off Greenlandic glaciers and drift on paths that are notoriously hard to predict, polar lows can deepen faster than almost any mid-latitude storm, and congestion data from the high Arctic often arrives too late to be useful. We built AROS to tackle all of this at once rather than piece by piece. It is a browser-only voyage planning tool, written in React and TypeScript, that chains four ML inference modules into a single workflow: a TensorFlow.js neural network scoring route risk from a ten-dimensional environmental feature vector; a Random Forest regressor estimating weather from location and departure date; a K-Means model mapping active vessel positions into congestion zones; and a binary network outputting iceberg collision probabilities for individual ship-iceberg pairs. Ten ports across Arctic Russia, Norway, Greenland, Canada, and Antarctica are supported. Waypoints are interpolated along geodetic arcs and validated against sea-region bounding boxes to keep routes over water throughout. Testing across all supported port pairings confirmed correct risk classification over the full range of ice concentration and iceberg count values, with every computation finishing inside a single browser render cycle. The outcome shows that a client-only architecture is sufficient for operationally useful Arctic maritime decision support.
Keywords: Arctic maritime routing, sea ice concentration, iceberg col-lision risk, Random Forest, K-Means clustering, TensorFlow.js, OpenLayers, route optimization, vessel traffic analysis.
Keywords: Arctic maritime routing, sea ice concentration, iceberg col-lision risk, Random Forest, K-Means clustering, TensorFlow.js, OpenLayers, route optimization, vessel traffic analysis.
Abstract
Study of Design and Mobility Approaches in Rocker–Bogie Suspension Systems
Sarthak Sharma, Utkarsh Nehwal, Yash Kaushik, Shrikant, Sushant Dhasmana, Dr. Uruj Jaleel, Dr. Satish Kumar Soni
DOI: 10.17148/IJARCCE.2026.154169
Abstract: Navigating rugged, unpredictable terrain is a significant engineering challenge for both off-road vehicles on Earth and rovers on Mars. When the ground becomes extremely rough, traditional suspension systems typically have trouble keeping a car stable and maintaining traction. The rocker-bogie suspension is useful in this situation. It successfully resolves these issues by maintaining all wheels in continuous contact with the ground, which makes it considerably simpler to crawl over big obstacles without tipping. It was first created for space missions. The rocker-bogie mechanism is thoroughly examined in this work. We examine its background, the workings of its mechanics, important design decisions, and the specifics of how it safely climbs over obstacles. Lastly, we examine the practical applications of utilizing this space-age technology.
Keywords: Rocker–bogie suspension, rover mobility, rough terrain navigation, off-road vehicles, suspension systems.
Keywords: Rocker–bogie suspension, rover mobility, rough terrain navigation, off-road vehicles, suspension systems.
Abstract
TradeSphere: A Full Stack Web-based Stock Trading Platform Using MERN Stack
Aaditya Gupta, Akshay Dhiman, Aman Kumar, Harsh Sharma, Dr.Uruj Jaleel, Dr.Satish Kumar Soni
DOI: 10.17148/IJARCCE.2026.154170
Abstract: The rapid expansion of financial markets and the increasing participation of students, young professionals, and first-time investors in stock trading have created a strong demand for educational tools that simplify the understanding of market operations. In recent years, digital trading platforms have become highly popular due to the availability of smartphones, internet access, and low-cost investment options. However, most existing trading applications are designed for professional traders and real investors, making them complex and difficult for beginners to understand. These platforms generally include advanced trading charts, multiple financial indicators, margin calculations, technical analysis tools, and live market data, which may overwhelm new learners. As a result, there is a growing requirement for a simplified educational trading system that can help users understand the core concepts of stock market trading without involving real financial risk.
This paper presents TradeSphere, a simulation-based full-stack web application developed to provide an easy, structured, and accessible environment for learning the basic functionalities of a stock trading platform. The system has been designed mainly for students, academic demonstrations, and beginner-level users who want to understand how modern online trading systems work. Instead of interacting with actual stock exchanges or real money transactions, TradeSphere simulates the major components of a stock trading dashboard, including stock holdings, market positions, order placement, and portfolio management.
Keywords: TradeSphere, MERN Stack, Stock Trading Platform, React.js, Node.js, Express.js, MongoDB, REST API, Portfolio Management, Trading Dashboard.
This paper presents TradeSphere, a simulation-based full-stack web application developed to provide an easy, structured, and accessible environment for learning the basic functionalities of a stock trading platform. The system has been designed mainly for students, academic demonstrations, and beginner-level users who want to understand how modern online trading systems work. Instead of interacting with actual stock exchanges or real money transactions, TradeSphere simulates the major components of a stock trading dashboard, including stock holdings, market positions, order placement, and portfolio management.
Keywords: TradeSphere, MERN Stack, Stock Trading Platform, React.js, Node.js, Express.js, MongoDB, REST API, Portfolio Management, Trading Dashboard.
Abstract
Land Use and Land Cover Classification Using Sentinel-2 Satellite Imagery
𝐏𝐫𝐨𝐟. 𝐀𝐦𝐢𝐭 𝐌𝐞𝐬𝐡𝐫𝐚𝐦, 𝐃𝐡𝐚𝐧𝐬𝐡𝐫𝐢 𝐃𝐮𝐤𝐚𝐫𝐞, 𝐒𝐚𝐧𝐠𝐡𝐚𝐫𝐚𝐭𝐧𝐚 𝐏𝐚𝐭𝐢𝐥, 𝐑𝐢𝐭𝐞𝐬𝐡 𝐋𝐨𝐧𝐚𝐫𝐞
DOI: 10.17148/IJARCCE.2026.154171
Abstract: Land Use and Land Cover (LULC) classification is an important task in remote sensing for environmental monitoring, agriculture, and urban planning. This paper presents a deep learning-based approach for classifying satellite images into different land cover categories using a Convolutional Neural Network (CNN). The model is trained on the EuroSAT dataset, which consists of Sentinel-2 satellite images categorized into 10 classes such as forest, residential, river, and agricultural land. The proposed model uses multiple convolutional layers along with batch normalization and dropout to improve performance and reduce overfitting. Experimental results show that the model achieves high accuracy and performs effectively in distinguishing different land cover types. This system can be used for real-world applications such as land monitoring and disaster management.
Keywords: LULC, CNN, Deep Learning, Satellite Imagery, EuroSAT, Remote Sensing
Keywords: LULC, CNN, Deep Learning, Satellite Imagery, EuroSAT, Remote Sensing
Abstract
Smart Traffic Management System Using Machine Learning for Adaptive Signal Control and Emergency Vehicle Prioritization
Aditya Nanote, Srujal Ogale, Vikrant Palvi, Lokesh Patil, Prof. Rakesh C. Suryawanshi
DOI: 10.17148/IJARCCE.2026.154172
Abstract: The rapid growth of urban populations and vehicle density has resulted in severe traffic congestion, increased travel time, and environmental pollution in modern cities. Conventional traffic control systems rely on fixed signal timings or reactive approaches, which are inefficient in handling dynamic and unpredictable traffic conditions. This paper proposes a Smart Traffic Management System (STMS) that leverages machine learning and data-driven techniques to optimize traffic flow. The system employs a Random Forest Regression model to predict future traffic density based on historical datasets and simulated scenarios, analysing parameters such as vehicle count, time intervals, and lane-wise traffic patterns. An intelligent decision-making module dynamically adjusts signal timings based on these predictions. The system additionally incorporates an emergency vehicle prioritization mechanism and a visualization module providing analytical insights into traffic trends and prediction accuracy. By eliminating costly IoT infrastructure and focusing on predictive analytics, the proposed system offers a cost-effective, scalable, and flexible solution for modern traffic management, achieving over 96% packet delivery reliability and end-to-end latency as low as 200-350 ms under Wi-Fi conditions.
Keywords: Adaptive Signal Control, Emergency Vehicle Prioritization, IoT, Machine Learning, Predictive Analytics, Random Forest Regression, Smart Traffic Management.
Keywords: Adaptive Signal Control, Emergency Vehicle Prioritization, IoT, Machine Learning, Predictive Analytics, Random Forest Regression, Smart Traffic Management.
Abstract
SIGN LANGUAGE ANALYSIS USING ARTIFICIAL INTELIGENCE
Aarti Dhage, Sanika Pawar, Vaishnavi Tilekar, Kanyakumari Mangrule, Prof. Dr. Sachin Bere, Prof. Mr. A.M Suryawanshi
DOI: 10.17148/IJARCCE.2026.154173
Abstract: This paper introduces “SOUL” Sign Language Understanding and Learning, a real-time sign language detection and translation system that aims to fill the communication gap between hearing-impaired people and non- signers. The proposed system uses a hybrid machine learning approach, which combines Convolutional Neural Networks (CNN) for feature extraction and Random Forest Classifiers for gesture recognition. The system uses computer vision, natural language processing (NLP), and text-to-speech (TTS) techniques to enable seamless bidirectional translation between sign language and spoken language. The proposed system has an accuracy of 94.2% in recognizing American Sign Language (ASL) alphabets and phrases, with a real-time processing speed of 18 FPS. Moreover, the system is capable of multilingual translations (English and Marathi), making it flexible for use in different linguistic settings.
Keywords: Sign Language Recognition, CNN, Random Forest, Real-Time Translation, TTS, Accessibility.
Keywords: Sign Language Recognition, CNN, Random Forest, Real-Time Translation, TTS, Accessibility.
Abstract
Optimized Travel Recommendation & Itinerary Generation
Nidhi Dhake, Shravani Mestry, Akshay Diwate, Nishant Gudade, Dr. Siddharth Hariharan
DOI: 10.17148/IJARCCE.2026.154174
Abstract: The rapid growth of digital travel platforms has created an urgent need for intelligent, personalized travel recommendation systems that go beyond simple keyword-based search. This paper presents TripSona, a full-stack AI- powered Indian travel recommendation and itinerary planning system built on a microservices architecture. The system employs a hybrid recommendation engine that combines a trained Machine Learning model using collaborative and content-based filtering with real-time data from Google Places API and Booking.com API to generate personalized destination recommendations. For itinerary generation, the system introduces a novel Multi-Agent Architecture comprising six specialized Gemini AI agents — Day Planner, Weather, Budget, Food and Tips, Transport, and Coordinator — each responsible for a distinct aspect of trip planning. The system is trained and evaluated on a dataset of 2000 Indian travel records spanning 40 destinations across diverse categories including Beach, Adventure, Nature, Historical, and City. User preferences including age range, budget, trip duration, interests, purpose, health conditions, and cuisine preference are used as input features. Experimental results demonstrate that the hybrid recommendation approach achieves superior personalization accuracy compared to standalone ML or API-based approaches. The multi- agent itinerary system produces contextually rich, weather-aware, budget-conscious day-by-day travel plans with real venue data. The system is deployed as a React-based web application integrated with Firebase Authentication and PostgreSQL for persistent storage.
Keywords: Travel Recommendation System, Multi-Agent Architecture, Machine Learning, LangChain, Gemini AI, Google Places API, Microservices, Collaborative Filtering, Content-Based Filtering, Indian Tourism
Keywords: Travel Recommendation System, Multi-Agent Architecture, Machine Learning, LangChain, Gemini AI, Google Places API, Microservices, Collaborative Filtering, Content-Based Filtering, Indian Tourism
Abstract
SPARK-A Dating Application: A JAVA, Flutter and ANDRIOD-Based
Prof. Madhuri Parate, Ishant Fulzele, Yash Jugseniya, Gaurav Katole, Aditya Rewatkar, Rutuja Bhende
DOI: 10.17148/IJARCCE.2026.154175
Abstract: The rapid growth of mobile applications has significantly transformed the way people connect and communicate. This project, titled “Spark – A Modern Cross-Platform Dating Application,” focuses on developing a user-friendly and efficient dating platform using Flutter and Firebase technologies. The application is designed to operate seamlessly across multiple platforms, including Android, iOS, and Web, using a single codebase.
The primary objective of the project is to provide a simple, secure, and real-time environment for users to discover potential matches, express interest through a swipe-based interface, and communicate instantly upon mutual matching. The system incorporates features such as user authentication using email and Google OAuth, profile creation and management, swipe-to-like functionality, automatic match detection, and real-time chat powered by Cloud Firestore.
Keywords: Flutter, Firebase, Cross-Platform Application, Real-Time Chat, Dating App, Cloud Firestore, User Authentication, Social Networking, Mobile Application Development.
The primary objective of the project is to provide a simple, secure, and real-time environment for users to discover potential matches, express interest through a swipe-based interface, and communicate instantly upon mutual matching. The system incorporates features such as user authentication using email and Google OAuth, profile creation and management, swipe-to-like functionality, automatic match detection, and real-time chat powered by Cloud Firestore.
Keywords: Flutter, Firebase, Cross-Platform Application, Real-Time Chat, Dating App, Cloud Firestore, User Authentication, Social Networking, Mobile Application Development.
Abstract
AN OVERVIEW ON: CREDIT RISK ANALYSIS
Prof. Amit Meshram, Bhagyashree Kohad, Saloni Chitalkar, Pranali Ganvir, Nikita Adhau, Himanshu Tadas
DOI: 10.17148/IJARCCE.2026.154176
Abstract: Credit risk analysis is essential for evaluating the likelihood of borrowers defaulting on loans. This study uses historical financial and customer data to develop models that assess creditworthiness. Various factors such as income, credit history, and repayment behavior are analyzed to identify risk patterns. Statistical and machine learning techniques are applied to improve prediction accuracy. The findings highlight the importance of data-driven approaches in minimizing financial risk and supporting effective lending decisions.
Keywords: Credit Risk, Creditworthiness, Default Prediction, Machine Learning, Financial Analysis, Risk Assessment, Logistic Regression, Data Analysis
Keywords: Credit Risk, Creditworthiness, Default Prediction, Machine Learning, Financial Analysis, Risk Assessment, Logistic Regression, Data Analysis
Abstract
Healthcare Android Predictor
Ashwini Vatne, Pallavi Shrimangle, Dipali Lahane, Prof. D. S. Jaybhay
DOI: 10.17148/IJARCCE.2026.154177
Abstract: This project is an intelligent healthcare app for Android that makes medical services easier to use, more efficient, and more engaging for users. The app uses machine learning, location services, and AI-driven chatbots to offer smart and easy healthcare support. One of its main features is the ability to predict diseases based on the symptoms users enter. A machine learning model looks at the information provided and gives possible health issues, helping users get a better understanding of their condition before seeing a doctor. To make the app more useful and easier to use, it has an AI-powered chatbot that gives quick health advice, supports voice communication, and can interact in Hindi for better reach. The location feature lets users find nearby doctors and hospitals, using data from registered medical professionals who can add and update their service locations in the system.The app has two user types: patients and doctors. Patients can sign up, create profiles, and use features like symptom-based disease prediction, chatbot help, and finding nearby healthcare services. Doctors can log in, update their profiles, and list their available services and locations. The system focuses on keeping data safe and offers real-time help, automatic health insights, and easy access to medical resources.
Keywords: Artificial Intelligence Healthcare App, Machine learning Disease Prediction, Ai Chatbot, Voice and Hindi Language Support, Location Based Doctor Finder, Patient and Doctor Management System, Real-time Healthcare Assistant.
Keywords: Artificial Intelligence Healthcare App, Machine learning Disease Prediction, Ai Chatbot, Voice and Hindi Language Support, Location Based Doctor Finder, Patient and Doctor Management System, Real-time Healthcare Assistant.
Abstract
Optimized Automatic Timetable Generation Using Genetic Algorithm for Educational Institutions
Yash Gadekar, Ganesh Gaikwad, Diksha Gaikwad, Prof. Veena Bhamre
DOI: 10.17148/IJARCCE.2026.154178
Abstract: Timetable creation in schools and colleges is a complex, time‑consuming and error‑prone manual task.Administrators must assign teachers, subjects, rooms and time slots while satisfying numerous constraints such as faculty availability, room capacity, and avoiding clashes between classes, which makes the problem NP‑hard (Chen et al., 2021; Bashab et al., 2022). Manual scheduling often leads to conflicts, under‑utilization of resources and difficulty in adapting to changes during the semester (N, 2025; Bashab et al., 2022; Rudová et al., 2011). To address these challenges, this work proposes an automatic timetable generator based on a Genetic Algorithm (GA). The system models hard constraints (no overlaps, capacity, mandatory breaks) and soft constraints (teacher preferences, minimum gaps, balanced workload) in a fitness function and iteratively evolves candidate timetables using selection, crossover, and mutation (Dhomne, 2025; Dave et al., 2025; Katkar, 2024; Singhal et al., 2024). A web‑based interface allows users to input teachers, subjects, rooms and constraints, visualize generated timetables and regenerate schedules when requirements change. Experimental evaluation on sample institutional data indicates a significant reduction in generation time, fewer conflicts and improved room and faculty utilization compared to manual methods (Dhomne, 2025; Dave et al., 2025; Katkar, 2024; Singhal et al., 2024). The proposed system demonstrates that GA‑based automatic timetabling can provide a practical, scalable and user‑friendly solution for educational scheduling.
Keywords: Timetable Generation, Genetic Algorithm, Scheduling, Optimization, Academic Automation, Constraint Handling
Keywords: Timetable Generation, Genetic Algorithm, Scheduling, Optimization, Academic Automation, Constraint Handling
Abstract
The Impact of AI on Financial Decision Making in The Automobile Industry
Dhanalakshmi G, Dr Lumina Julie R
DOI: 10.17148/IJARCCE.2026.154180
Abstract: The use of advanced Artificial Intelligence (AI) is quickly changing the way organizations in a wide variety of industries (including automotive) use financial decision-making processes to complete day today tasks. This research specifically looks at how AI technology is changing the way automobile companies make financial decisions, specifically regarding areas of financial forecasting, risk assessment, investment planning and cost efficiency. The primary goal of this research is to analyze how AI-based tools and data analytics can provide automobile companies with more accurate, productive and strategic financial decisions. In this study, I utilized a descriptive research methodology to conduct my analysis using both primary and secondary sources of data. The primary data was collected using a structured survey of finance professionals and managers from selected automotive companies and analysis of their corporate reports and literature in the area of AI and finance. Data analysis revealed that AI has greatly increased the accuracy of financial forecasts, reduced the incidence of operational risk, and increased the ability of companies' strategic planning functions to utilize AI technology in their strategic decision-making processes. I also found that AI provides companies with the ability to analyze large quantities of financial data at a relatively faster rate than what was previously available, providing companies with more timely and informed financial decisions.
Keywords: Financial decision making, Automobile Industry, Financial Forecasting, Data analytics, Risk management.
Keywords: Financial decision making, Automobile Industry, Financial Forecasting, Data analytics, Risk management.
Abstract
A Study on Hr Policy Development and Data Management in Biotechnology Sector
Jayakumar R, Dr Tamilselvan P
DOI: 10.17148/IJARCCE.2026.154181
Abstract: The biotechnology industry is a field of high-tech innovation, regulatory compliance, and a high-skilled workforce, making it essential to develop effective HR policies and manage employee data to achieve organizational success. This study aims to examine the process, challenges, and best practices associated with HR policy development and employee data management in biotechnology organizations. The study has been conducted to identify some of the important HR functions such as recruitment, training and development, performance management, compliance, employee retention, etc., which are essential to achieve organizational success while complying with biotechnology industry regulations. In addition to this, it also identifies the importance of data management systems to enhance organizational effectiveness through effective HR functions. The study has been conducted through a mixed approach of qualitative and quantitative research to identify how biotechnology organizations can effectively integrate HR policies with state-of-the- art data management systems to achieve organizational effectiveness. The study has identified that effective HR policy development, along with effective data management practices, plays a major role in employee satisfaction, organizational effectiveness, and compliance with industry regulations. The study concludes that effective HR functions can be achieved through effective HR policy development, implementation of digital HR systems, effective data management practices, etc.,
Keywords: Human Resource Management, Policy development, Employee data, Recruitment, Biotechnology sector, HR Analytics, Data security.
Keywords: Human Resource Management, Policy development, Employee data, Recruitment, Biotechnology sector, HR Analytics, Data security.
Abstract
Impact of Buy Now Pay Later Services on the Spending Behavior of Generation Z Consumers in Chennai
Karthikeyan S, Dr.Lumina Julie R
DOI: 10.17148/IJARCCE.2026.154182
Abstract: Buy Now Pay Later (BNPL) has undergone a huge transformation in how customers are making their purchases as compared to years past. Young consumers (Generation Z) especially prefer more flexible and technological financing options, but little empirical research has been conducted on how BNPL affects the way young consumers shop in metropolitan areas in India. Therefore, the purpose of this study is to collect and objectively study how BNPL affects how Generation Z consumers in Chennai shop using the specific areas of impulsive purchases, frequency of purchases, and financial knowledge. Using a convenience sampling approach, the study collected primary data from respondents in Chennai's Generation Z population through the use of an established questionnaire. The data analysis methods used were quantitative and included descriptive statistics, correlation, and regression analysis to test the relationship between having used BNPL and the different shopping characteristics measured.
The results of the research found that the use of BNPL had a significant positive influence on the impulsive purchasing and purchases made frequently by those who have used BNPL as well as reducing the perceived immediate financial burden. In total, however, by the results, consumers appear to have an average understanding of the repayment responsibilities that accompany this form of financing and the long-term effects of this type of financing on their financial situation.
Keywords: Buy Now Pay Later (BNPL), Generation Z Consumers, Spending Behavior, Impulsive Buying, Purchase Frequency, Financial Awareness
The results of the research found that the use of BNPL had a significant positive influence on the impulsive purchasing and purchases made frequently by those who have used BNPL as well as reducing the perceived immediate financial burden. In total, however, by the results, consumers appear to have an average understanding of the repayment responsibilities that accompany this form of financing and the long-term effects of this type of financing on their financial situation.
Keywords: Buy Now Pay Later (BNPL), Generation Z Consumers, Spending Behavior, Impulsive Buying, Purchase Frequency, Financial Awareness
Abstract
AI-Based Expense Tracker for Personal Finance Management
Kanishka, Prachi Pundir, Nandini Kumari, Mayank Kumar Roy, Uruj Jaleel, Satish Kumar Soni
DOI: 10.17148/IJARCCE.2026.154183
Abstract: Artificial Intelligence (AI) is rapidly changing the field of education by making learning more effective, personalized, and accessible. This paper discusses how AI technologies such as machine learning, virtual classrooms, and intelligent tutoring systems are improving the quality of education. AI helps students learn at their own pace, provides instant feedback, and supports teachers by automating routine tasks like grading and attendance.
The study also highlights the various applications of AI in education, including personalized learning, language translation, and performance analysis. In addition, AI makes education more inclusive by helping students from different backgrounds and abilities.
However, there are also some challenges, such as high implementation cost, data privacy issues, and lack of proper training for teachers. Despite these challenges, AI has a strong potential to transform the future of education. Overall, the paper concludes that teachers when used responsibly.
The study also highlights the various applications of AI in education, including personalized learning, language translation, and performance analysis. In addition, AI makes education more inclusive by helping students from different backgrounds and abilities.
However, there are also some challenges, such as high implementation cost, data privacy issues, and lack of proper training for teachers. Despite these challenges, AI has a strong potential to transform the future of education. Overall, the paper concludes that teachers when used responsibly.
Abstract
DIGITALIZATION OF HR OPERATIONS IN THE SECURITY SERVICES SECTOR
Nithish K R, Dr. Geeta Kesavaraj
DOI: 10.17148/IJARCCE.2026.154184
Abstract: This research is about the impact of the changing nature of human resource operations in the security services sector that is brought about by digitalization. It looks couple of company with lot of employees. The study wants to know how employees feel like using tools for things like getting paid checking attendance and talking to each other. The people who did the study used a design to collect data from 124 employees in different parts of the company. They used a questionnaire for asking questions. Then used special tools to analyze the answers. What they did discover is that even though digital tools are being used a lot employees are not totally on board. Most employees said they were not sure if digital tools were really making things more efficient or helping them to communicate better. Some employees said that they did not get training for using the digital tools and had problems in using them. The study further discovered that younger employees had to be more willing to use tools compared to the older ones. This indicates that there is a difference in the way different age groups adapt to technological change. The study says that digitalization can really help Human Resource operations but only if the digital tools are easy to use and employees get the training they need. The people who did the study think that companies should deliver hands-on training as well as ensure the digital tools are not too hard to use on phones.
Keywords: HR Digitalization, Perception and Efficiency of employees, Training and Adoption security services.
Keywords: HR Digitalization, Perception and Efficiency of employees, Training and Adoption security services.
Abstract
Impact of Corporate Social Responsibility on Community well-being among IT Industry.
Prabha C, Dr.Lumina Julie R
DOI: 10.17148/IJARCCE.2026.154185
Abstract: Corporate Social Responsibility (CSR) has grown to be a strategic tool for sustainable development globally, especially in emerging economies like India where CSR spending has been legally made compulsory through the Companies Act, 2013. Our research focuses on how CSR initiatives impact the community's well, being in the IT sector located in Tamil Nadu. Research studied four facets of CSR, environmental responsibility, social welfare programs, ethical business practices, and employee, centered CSR with reference to community well, being. It adopted a quantitatively research approach through the use of structured questionnaires. Out of the 384 surveys that were distributed, 116 completed ones were verified and analyzed using SPSS. The consistency and reliability of the findings were confirmed through the measures of Cronbachs alpha, KMO as well as Bartletts test. Analyse used multiple regression analysis to determine the relationship between the variables. The results reveal that both environmental responsibility and social welfare programs have a major effect on community well, being, whereas ethical business practices and employee, centered CSR have shown a moderate positive impact. The research work enriches CSR literature by presenting empirical evidence from the Indian IT industry with sectoral focus and also provides suggestions for policymakers and corporate leaders who are looking to create impactful CSR strategies.
Keywords: Corporate Social Responsibility, Community Well-Being, IT Industry, Tamil Nadu, Environmental Responsibility, Social Welfare Programs
Keywords: Corporate Social Responsibility, Community Well-Being, IT Industry, Tamil Nadu, Environmental Responsibility, Social Welfare Programs
Abstract
A STUDY ON LOGISTICS OPTIMIZATION ON MILK COLLECTION AND DISTRIBUTION
R. Dhanush, Kokila.K
DOI: 10.17148/IJARCCE.2026.154186
Abstract: This paper is about streamlining logistical operations in the milk collecting and distribution network of the Andaman and Nicobar Islands Integrated Development Corporation (ANIIDCO). This is aimed at achieving efficiency, minimization of operation cost, and delivery of fresh milk in time. The study examines route planning, cold chain maintenance, fuel consumption, and technology application as the main factors among the supply chain farmers, collection agents, transporters, and distribute with the help of a descriptive research design and a purposive sample of 116 respondents throughout the supply chain. Structured questionnaires were used to collect the data which were analyzed in terms of percentage analysis, regression, chi-square, and t-tests. The results indicate that although route planning and real time tracking have enhanced the operations, there are still some challenges in the areas of communication, vehicle maintenance and digital tools. Poor implementation of technology and erratic scheduling have impacts on quality of milk and expenses. The study concludes by stating that a combination of GPS tracking, data analytics, and efficient communication systems will be able to enhance the performance of logistics to a considerable extent. Some of the recommendations would be to optimize routes better, train more, and service equipment. ANIIDCO will be able to create an efficient, cost effective, and sustainable milk supply chain to the island region by enhancing coordination and embracing digital innovation.
Keywords: Technology utilization, Cold chain maintenance, Sustainable milk supply chain.
Keywords: Technology utilization, Cold chain maintenance, Sustainable milk supply chain.
Abstract
AN OVERVIEW ON: CREDIT RISK ANALYSIS
Prof. Madhuri Parate, Meena Godghate, Hanisha Bulhe
DOI: 10.17148/IJARCCE.2026.154187
Abstract: Credit risk analysis is essential for evaluating the likelihood of borrowers defaulting on loans. This study uses historical financial and customer data to develop models that assess creditworthiness. Various factors such as income, credit history, and repayment behavior are analyzed to identify risk patterns. Statistical and machine learning techniques are applied to improve prediction accuracy. The findings highlight the importance of data-driven approaches in minimizing financial risk and supporting effective lending decisions.
Keywords: Credit Risk, Creditworthiness, Default Prediction, Machine Learning, Financial Analysis, Risk Assessment, Logistic Regression, Data Analysis
Keywords: Credit Risk, Creditworthiness, Default Prediction, Machine Learning, Financial Analysis, Risk Assessment, Logistic Regression, Data Analysis
Abstract
MANAGING STRESS IN THE WORKPLACE THROUGH HR PRACTICES IN IT INDUSTRY.
Robina A, Dr. R. Lumina Julie
DOI: 10.17148/IJARCCE.2026.154188
Abstract: Work-related stress arising from long working hours, heavy workloads, tight deadlines, and the need to continuously update one's knowledge with new technologies is increasing rapidly in the IT sector. In such situations, workers get exposed to high levels of stress which causes their mental health to deteriorate and organizational effectiveness to decline. As a result, it is imperative to determine the HR practices which help in managing and reducing workplace stress. This study aims primarily to explore the link between work, related stress and chosen HR practices in the IT sector. It centres around three main aspects: working, life balance, job autonomy, and workload. Working, life balance is about a person's ability to cope efficiently with work and personal life, job autonomy deals with how much freedom employees have in deciding the way they carry out tasks, and workload is the amount of work and duties that a person is expected to undertake.
The methodology for this research is mostly quantitative, and data from employees of IT companies will be gathered by way of a survey. Afterward, the data will be processed through factor analysis, regression analysis, and correlation tests to discover the association between HR practices and stress levels. This research's results can be of great use to HR managers in: shaping well, targeted policies and programs aiming at alleviating employee stress, enhancing work satisfaction, boosting work performance, and building a green work environment that IT industry employees can thrive in.
Keywords: Workplace Stress, Work-Life Balance, Job Autonomy, Workload, HR Practices, IT Industry
The methodology for this research is mostly quantitative, and data from employees of IT companies will be gathered by way of a survey. Afterward, the data will be processed through factor analysis, regression analysis, and correlation tests to discover the association between HR practices and stress levels. This research's results can be of great use to HR managers in: shaping well, targeted policies and programs aiming at alleviating employee stress, enhancing work satisfaction, boosting work performance, and building a green work environment that IT industry employees can thrive in.
Keywords: Workplace Stress, Work-Life Balance, Job Autonomy, Workload, HR Practices, IT Industry
Abstract
A Study on Impact of Social Media Marketing Strategies on Consumer Buying Behaviour
Ms S. Mithra, Dr. M. Rajapriya
DOI: 10.17148/IJARCCE.2026.154189
Abstract: This study basically tries to understand how social media marketing actually shapes the way People decide what to buy, especially with how fast digital media is growing in India. Instead Of looking at social media in a broad way, I focused more on the things people see every day Like reels, review videos, influencer posts, ads that pop up while scrolling, and small Engagement activities like polls or Q&A sessions. The idea was to see whether these things really change how people feel about a brand and if it eventually pushes them toward buying something.Problem I Noticed:While observing the online market, I kept seeing a few issues. For example, influencer Marketing has become so common that it’s honestly hard to tell who genuinely likes a Product and who is promoting it just for the collaboration. Another thing is the sheer number Of brands posting similar types of content. Sometimes consumers get confused because all the Posts start looking the same, and the decision becomes less about “Is the product good?” and More about “Is this trending right now?” So I wanted to figure out how much of this actually Affects real buying behaviour. I used a descriptive research design because I mainly wanted to Understand people’s current opinions and behaviour instead of experimenting with anything. In the model, Social Media Marketing Factors were treated as the independent variables. These include the usual things like content type, ads, influencer posts, etc.These factors Influence consumer buying behaviour (the dependent variable), but not directly. They work Through four mediator Brand Awareness, Brand Trust, Engagement, and Purchase Intention. Basically, if someone sees a post, it first makes them aware, then maybe trust develops, then They engage with it, and only after all that they think of actually buying.I collected data using a Google Forms questionnaire. Out of the first 200 responses, only 118 Were actually useful and valid. Most of the respondents were young adults aged 21 30, which Honestly makes sense because they are the most active online. For the analysis, I used Descriptive stats, correlation, and regression/ANOVA to understand the strength of the Relationships. One of the first interesting things that came out was platform usage. Even Though Instagram feels like the most happening app, Facebook still had the highest usage in My sample (39.8%), followed by Instagram (29.4%) and YouTube (24.7%). I didn’t expect Facebook to dominate, but maybe people still use it more quietly for groups, news, or Marketplace. Another part of the study involved testing whether the platform someone uses Affects how effective they think social media marketing is. The p-value was 0.120, which is Higher than 0.05, so statistically, it doesn’t have significance. In simple terms, the type of Platform doesn’t really matter people don’t rate SMM as effective or ineffective based on Whether they use Facebook, Instagram, or YouTube.
Keywords: Social Media Marketing, Consumer Buying Behaviour, Brand Awareness.
Keywords: Social Media Marketing, Consumer Buying Behaviour, Brand Awareness.
Abstract
A STUDY ON IMPACT OF WEBSITE USER EXPERIENCE CONVERSION RATES
Seethaladevi. S, Dr. Felisiya.M
DOI: 10.17148/IJARCCE.2026.154190
Abstract: The user website experience is really important for businesses. It affects how many people do what you want them to do on your website. This is called the conversion rate. It is the number of people who visit your website and then do something like buy something sign up for something or fill out a form. If your website is easy to use and looks good people will like it more. They will stay on your website longer. Do more things. A good website has things like navigation loads fast works well on phones looks nice and has buttons that tell you what to do next. When a website is hard to use and does not look good people will leave. They will not come back. You will lose customers. This study looks at how the user website experience affects whether people do things on your website or not. It shows that making your website easy to use and nice to look at is very important for your business. The study says that if you make your website better people will trust you more and your business will do better. It also says that a good user website experience can help people decide to do things on your website. This can make a difference for businesses that operate online. The user website experience is important, for businesses because it helps people do things on your website.
Keywords: User Experience (UX) Website Usability Conversion Rate User Interface (UI) Customer Engage
Keywords: User Experience (UX) Website Usability Conversion Rate User Interface (UI) Customer Engage
Abstract
Car Accident Detection with Alert System
Gayatri Mali, Shruti Desai, Vaibhavi Mhaske, Chaitanya Sonawane, Prof. Mrs. Dhage T.S, Mr.Bhapkar A.S
DOI: 10.17148/IJARCCE.2026.154191
Abstract: The accidents and the accident injuries in the world is increasing in our day today life so there must be good and efficient control for the safety of human life violation of traffic rules drunk driving, careless driving are because of road accidents as we know we cannot stop the accidents but we can reduce the accidents by some precautionary measures. Road accident is most undesirable thing to happen to a road user, though they happen quite often. Most of the fatal accidents happened due to over speeding. This system provides a good and securable driving for the driver. In this project we are controlling the speed, seat belt, obstacle detection, alert indication, notification about accident whole security of the vehicle. Proposed method resolves the problems faced by the existing systems. In this project the vehicle is partially controlled by various sensors. There is a vehicle black box it will store the data of various sensors which measure various performance of the vehicle and also sends information to various authorities about the vehicle position to get immediate first aid. When any abnormalities from the threshold value occurs, the microcontroller sends message to the desired positions and immediate first aid and other facilities are available. This system is suitable in all type of vehicles.
Keywords: IoT, Accident Detection, GSM, GPS, Arduino, Vehicle Safety, Real-Time Monitoring.
Keywords: IoT, Accident Detection, GSM, GPS, Arduino, Vehicle Safety, Real-Time Monitoring.
Abstract
A STUDY ON SEO STRATEGY DEVELOPMENT AND IMPLEMENTATION FOR SMALL BUSINESSES: A CASE STUDY OF WEBOIN TECHNOLOGIES PVT LTD
Thangavelan K, Dr. Felisiya. M
DOI: 10.17148/IJARCCE.2026.154192
Abstract: Small businesses in modern times depend on Search Engine Optimization (SEO) which serves as their main method for improving digital visibility and competing against other businesses. Small businesses need more digital presence because their resources are limited and their SEO efforts produce inadequate results. This research paper explores how search engine optimization strategy development affects online business performance for small businesses through its examination of Weboin Technologies Pvt Ltd. The study aims to explore how SEO activities which include keyword research and on-page optimization and technical SEO and content strategy development affect search engine rankings and organic traffic growth and lead generation results. The paper uses existing literature and digital marketing theoretical frameworks to create its conceptual approach. The results demonstrate that using appropriate SEO methods enables businesses to achieve successful online marketing and reach their target audience. The paper demonstrates that businesses should develop structured SEO plans because they will help establish brand identity and drive business development in the highly competitive digital market.
Abstract
A STUDY ON SOCIAL MEDIA MARKETING AND THE IMPACT OF ONLINE PROMOTION WITH REFERENCE TO WESTSIDE
M. Udheepa Reddy, Dr. Felisiya. M
DOI: 10.17148/IJARCCE.2026.154193
Abstract: Along the digital age, social media marketing has emerged as one of the most efficient means through which organizations can communicate with clients, and their products and services. Social media networks are gradually playing a crucial role in allowing retail companies to generate brand awareness, interact with customers and manipulate the decision-making process when it comes to purchase decisions. The reason why this conceptual research paper will examine this aspect is to investigate the importance of social media marketing and its influence on online promotion strategies within the retail sector using Westside as a reference. The case developed is aimed at comprehending the impact of social media activities like brand communication, campaign activities, online advertisements, and content usage in social media on online promotions such as digital offers, sponsored campaigns, and promo deals. The approach of the paper is conceptual, based on the survey of the existing literature and theoretical frameworks of digital marketing and consumer engagement. The data indicates that the right social media marketing techniques can make the promotion process on the Internet highly efficient and increase customer communication. This paper also emphasizes the need to incorporate social media marketing with on-line promotion strategies in enhancing brand recognition and customer relationships in the crowded retail market.
Keywords: Social Media Marketing, Online Promotion, Digital Marketing, Customer Engagement, and Retail Marketing.
Keywords: Social Media Marketing, Online Promotion, Digital Marketing, Customer Engagement, and Retail Marketing.
Abstract
Fruit Detection and Classification Using CNN with Deep Transfer Learning
Prerna, Sejal Rana, Dr. Satish Kumar Soni
DOI: 10.17148/IJARCCE.2026.154194
Abstract: Accurate and efficient fruit detection and classification are critical in modern precision agriculture, supply chain automation, and quality control. Manual inspection methods are time-consuming, error-prone, and impractical at scale. This paper proposes a deep learning framework that integrates Convolutional Neural Networks (CNN) with deep transfer learning techniques to automatically detect and classify multiple fruit categories from images. We fine-tune three pre-trained models — VGG19, ResNet-50, and MobileNetV2 — on the publicly available Fruits-360 dataset consisting of 90,380 images across 131 fruit classes. Data augmentation strategies including random flipping, rotation, and brightness adjustment were applied to improve model generalization. Experimental results demonstrate that the fine- tuned ResNet-50 model achieves the highest classification accuracy of 98.74%, while MobileNetV2 provides the best trade-off between accuracy and computational efficiency at 97.91% accuracy with significantly reduced inference time. The proposed approach outperforms several existing methods and shows strong potential for real-world agricultural deployment.
Keywords: Fruit Detection, Image Classification, Convolutional Neural Networks, Transfer Learning, Deep Learning, VGG19, ResNet-50, MobileNetV2, Fruits-360.
Keywords: Fruit Detection, Image Classification, Convolutional Neural Networks, Transfer Learning, Deep Learning, VGG19, ResNet-50, MobileNetV2, Fruits-360.
Abstract
SAHA -Women Safety Application and Device
Vedangi Shastrakar, Anamika Shirke, Sanika Shirsat, Parita Thakur, Dr. M.M. Deshpande
DOI: 10.17148/IJARCCE.2026.154195
Abstract: Women’s safety has become a major concern in today’s society due to the increasing number of harassment and emergency incidents occurring in both public and private spaces. Many existing safety solutions depend heavily on internet connectivity, which may not always be available during emergencies. To address this issue, this paper presents SAHA, a smart women safety system that combines both hardware and software components to provide immediate assistance and reliable monitoring.
The proposed system includes a portable hardware device along with a mobile application, ensuring functionality in both online and offline conditions. The hardware device is built using an ESP32 microcontroller and includes an SOS button, GSM/GPS module, and a buzzer. When the SOS button is pressed, the system sends emergency alerts along with the user’s location to pre-registered guardians through SMS, even without internet access. The buzzer also generates a loud sound to alert nearby people and discourage potential threats.
The mobile application further enhances the system by offering features such as live location tracking, safe zone detection, nearby emergency services, and guardian management. Additional features like fake call, voice-activated SOS, and safety awareness resources help users handle uncomfortable or risky situations effectively.
By integrating hardware reliability with intelligent software features, the proposed system provides fast response, improved safety awareness, and user confidence. The solution is portable, affordable, and easy to use, making it suitable for students, working professionals, and travelers. Overall, SAHA aims to improve personal security by offering both emergency support and preventive safety features.
Keywords: Women Safety, ESP32, GSM/GPS, SOS System, Mobile Application, Voice Detection, Safe Zones, Emergency Communication, IoT
The proposed system includes a portable hardware device along with a mobile application, ensuring functionality in both online and offline conditions. The hardware device is built using an ESP32 microcontroller and includes an SOS button, GSM/GPS module, and a buzzer. When the SOS button is pressed, the system sends emergency alerts along with the user’s location to pre-registered guardians through SMS, even without internet access. The buzzer also generates a loud sound to alert nearby people and discourage potential threats.
The mobile application further enhances the system by offering features such as live location tracking, safe zone detection, nearby emergency services, and guardian management. Additional features like fake call, voice-activated SOS, and safety awareness resources help users handle uncomfortable or risky situations effectively.
By integrating hardware reliability with intelligent software features, the proposed system provides fast response, improved safety awareness, and user confidence. The solution is portable, affordable, and easy to use, making it suitable for students, working professionals, and travelers. Overall, SAHA aims to improve personal security by offering both emergency support and preventive safety features.
Keywords: Women Safety, ESP32, GSM/GPS, SOS System, Mobile Application, Voice Detection, Safe Zones, Emergency Communication, IoT
Abstract
Secure Image & Audio Steganography Using AES-256 Encryption and Hybrid DWT-SVD Frequency Domain Embedding
M. Tejas Srinivasan, S. Roshan Pranao, Y. Sai Dheeraj, Ms. N. Saranya
DOI: 10.17148/IJARCCE.2026.154196
Abstract: This paper presents a robust dual-layer secure image steganography framework that integrates AES-256 encryption in Counter (CTR) mode with a hybrid Discrete Wavelet Transform and Singular Value Decomposition (DWT- SVD) technique for covert communication. Unlike fragile spatial-domain methods such as Least Significant Bit (LSB) substitution, the proposed system operates entirely within the frequency domain by first encrypting the secret image using AES-256 with SHA-256 key derivation, and then embedding the resulting ciphertext into the high-frequency sub-bands of the cover image's luminance (Y) channel in the YCbCr colour space. Experimental evaluation demonstrates a Peak Signal-to-Noise Ratio (PSNR) of 55.58 dB between cover and stego images, a Structural Similarity Index (SSIM) of 0.99998, a Mean Squared Error (MSE) of 0.1797, and 100% Bit Correct Recovery (BCR) under ideal conditions. The system is delivered as three complementary user interfaces: a Tkinter desktop GUI, a Streamlit web dashboard, and a command-line interface (CLI). An additional audio steganography module extends the same AES-CTR pipeline to WAV files via LSB embedding, demonstrating the modularity of the design. The work addresses key gaps identified in the existing literature, namely the absence of a combined cryptographic and frequency-domain embedding approach, and establishes a publicly benchmarked baseline for future research into error-correcting codes, adaptive alpha tuning, and deep learning steganalysis resistance.
Keywords: Steganography, AES-256, DWT-SVD, Frequency-Domain Embedding, Image Security, PSNR, SSIM, Covert Communication, YCbCr, Cryptography.
Keywords: Steganography, AES-256, DWT-SVD, Frequency-Domain Embedding, Image Security, PSNR, SSIM, Covert Communication, YCbCr, Cryptography.
Abstract
Leveraging AI for the Next Era of Precision Oncology in Breast Cancer Detection
Amritpal Singh Yadav*, Virendra Kumar Sharma
DOI: 10.17148/IJARCCE.2026.154197
Abstract: Cutting-edge progress in ML and DL techniques has led to substantial gains in the accuracy and reliability of breast cancer diagnostic systems. Conventional diagnostic means could have limited sensitivity and can be subjective; therefore, AI supported Computer-Aided Diagnosis (CAD) systems are introduced to overcome these issues. This review summarizes developments from 2022–2025 in ML and DL techniques applied to mammography, ultrasound, MRI, histopathology, and thermography. Deep learning methods, with convolutional neural networks at the forefront, have achieved notable accuracy across multiple imaging types, while emerging trends such as radiomics, deep reinforcement learning, hybrid ML–DL frameworks, and explainable AI (XAI) further enhance diagnostic performance and clinical trust. Challenges including data scarcity, model interpretability, and generalization remain, with promising solutions found in self-supervised learning, federated learning, and foundation models. These advancements collectively support earlier detection, improved treatment planning, and the advancement of precision oncology.
Keywords: Breast cancer detection, machine learning, deep learning, CNN, CAD systems, radiomics, explainable AI, federated learning
Keywords: Breast cancer detection, machine learning, deep learning, CNN, CAD systems, radiomics, explainable AI, federated learning
Abstract
AshaConnect: Bridging Healthcare to Every Doorstep
Prapti Borkar, Sakshi Bagul, Tanmay Ghodke, Bhavesh Dhangar
DOI: 10.17148/IJARCCE.2026.154198
Abstract: Healthcare provision in rural India depends on the activities of Accredited Social Health Activists (ASHA workers), who traditionally utilize paper-based workflows. These workflows were found to be associated with frequent loss of information, travel issues, and poor patient show rates in referral hospitals. This research presents an innovative approach to digitize the process. This approach involves the design and implementation of the web-based health management system, AshaConnect. The system includes role-based portals for ASHA workers and hospitals, thus allowing for electronic patient registration, longitudinal tracking of health status, online appointment scheduling, and automatic exchange of reports between two parties. The solution has been built using technologies including Node.js, Express.js, SQLite, and web-native programming. It was built with three-tier RESTful architecture. The new software has been evaluated against the existing paper-based workflow by comparing the performance of both systems in eight aspects. The results showed that AshaConnect can bring substantial improvement in record keeping, hospital readiness, and patient care coordination. Overall, the proposed solution appears to be able to replace ASHA worker workflow fully.
Keywords: ASHA worker; rural healthcare; digital health; patient management; appointment booking; hospital coordination; health records; Node.js; SQLite
Keywords: ASHA worker; rural healthcare; digital health; patient management; appointment booking; hospital coordination; health records; Node.js; SQLite
Abstract
Immersi-Lab: Empowering Disabled Learners Through Virtual Biology Simulations
Faheem Ahmad, Dev Verma, Tanya Verma, Harsh Sharma, Dr Uruj Jaleel, Dr.Satish Kumar Soni
DOI: 10.17148/IJARCCE.2026.154199
Abstract: Traditional biology laboratories systematically exclude differently-abled students — those with physical disabilities, mobility impairments, chronic health conditions, and sensory limitations — from hands-on science education. In India, with over 26.8 million persons with disabilities [6], the inaccessibility of physical laboratory infrastructure represents a critical educational equity failure. This paper presents ImmersiLab, an immersive Virtual Reality (VR) Biology Laboratory developed on Unity Engine LTS with the XR Interaction Toolkit and OpenXR framework [7], deployed on Meta Quest 2/3 [9]. ImmersiLab provides five fully interactive, curriculum-aligned biology experiment modules — Cell Observation, Organ Dissection, Osmosis, Microbiology, and Photosynthesis — each structured as a four-stage guided workflow: Introduction, Preparation, Execution, and Review. By enabling safe, self-paced, location- independent, and infinitely repeatable biology experimentation, ImmersiLab removes every physical and systemic barrier faced by differently-abled learners. User Acceptance Testing conducted with 30 MCA students at MIET, Meerut demonstrated strong performance: 72+ FPS on Meta Quest 2, under 2GB RAM usage, app loading under 10 seconds, and user satisfaction averaging 4.5 out of 5 across all parameters. Research evidence confirms that VR-based learning produces 25% better retention than textbook-based methods [1], that virtual labs reduce costs by up to 60% [3], and that VR significantly reduces participation gaps for students with physical disabilities [4]. ImmersiLab is designed to democratize science education across India by making biology laboratory learning fully accessible regardless of physical ability or geographic location.
Keywords: Virtual Reality, Inclusive Education, Differently-Abled Learners, VR Biology Lab, Unity Engine, XR Toolkit, Accessibility, Meta Quest, Immersive Learning, India, Disability.
Keywords: Virtual Reality, Inclusive Education, Differently-Abled Learners, VR Biology Lab, Unity Engine, XR Toolkit, Accessibility, Meta Quest, Immersive Learning, India, Disability.
Abstract
A SURVEY OF AI-DRIVEN INTRUSION DETECTION SYSTEMS FOR CLOUD AND EDGE COMPUTING: TECHNIQUES, CHALLENGES, AND FUTURE DIRECTIONS
Abdulrahman Mohammed Saba, Alfa Muhammad, Sayuti Musa Shafi’i
DOI: 10.17148/IJARCCE.2026.154200
Abstract: The research presents a comprehensive survey of AI-driven intrusion detection systems designed for cloud and edge computing environments. The review systematically analyses recent research developments between 2015 and 2025, focusing on the application of supervised, unsupervised, semi-supervised, and hybrid learning techniques for network intrusion detection. It examines widely used algorithms such as Support Vector Machines, Random Forests, Convolutional Neural Networks, Recurrent Neural Networks, and emerging models including Transformer architectures and Graph Neural Networks. In addition, the survey evaluates commonly used benchmark datasets, such as NSL-KDD, CIC-IDS2017, and UNSW-NB15, which are widely employed to assess detection performance and model generalization. AI-driven intrusion detection systems represent a promising direction for strengthening cybersecurity in distributed cloud and edge computing ecosystems. By integrating advanced machine learning techniques with scalable and privacy-aware architectures, future IDS solutions can provide more intelligent, resilient, and proactive defence mechanisms against increasingly sophisticated cyber threats.
Keywords: Application, Algorithms, Models, Datasets.
Keywords: Application, Algorithms, Models, Datasets.
Abstract
AN OVERVIEW ON: CAMPUS HIRE
Prof. Frah Nikhat, Trupti Karemore, Shruti Ruikar, Umair Qureshi
DOI: 10.17148/IJARCCE.2026.154201
Abstract: The Campus Recruitment Management System (CRMS) is a web-based application designed to automate and streamline the placement process in educational institutions. The system creates a centralized platform where students, administrators, and recruiters can interact efficiently. Traditional recruitment methods rely heavily on manual record keeping, emails, and spreadsheets, which often lead to errors, delays, and miscommunication. The proposed system eliminates these issues by digitizing job postings, applications, eligibility verification, and result management.
Keywords: Campus Placement, Recruitment System, Student Portal, MERN Stack, Automation
Keywords: Campus Placement, Recruitment System, Student Portal, MERN Stack, Automation
Abstract
AI - DRIVEN INNOVATIONS IN MUTUAL FUND ANALYTICS
SANKAR C, Dr. S. ARUL KRISHNAN
DOI: 10.17148/IJARCCE.2026.154202
Abstract: This study looks at how Artificial Intelligence's changing the mutual fund sector. It focuses on ways Artificial Intelligence is being used between 2020 and 2025. The study checks how Artificial Intelligence can help with analyzing mutual fund performance looking at risks and understanding how investors behave. It does this by using computer models and predicting what will happen next. Artificial Intelligence uses computer codes, language understanding and looks at what people are saying about the market right now. This helps mutual fund managers make guesses about what will happen to mutual funds and the market. The study also looks at how Artificial Intelligence's helping to pick mutual funds automatically manage portfolios and give investors personalized advice. The results show that Artificial Intelligence has made changes in the mutual fund industry. It has made things more transparent, efficient. Helped investors make better decisions. Overall the study shows that Artificial Intelligence is changing the mutual fund industry by using data to make investment decisions and making the financial system smarter.
Keywords: Artificial Intelligence, in Mutual Funds, Predictive Analytics, Portfolio Optimization, Machine Learning, Investor Decision Support, FinTech Innovation.
Keywords: Artificial Intelligence, in Mutual Funds, Predictive Analytics, Portfolio Optimization, Machine Learning, Investor Decision Support, FinTech Innovation.
Abstract
A STUDY ON THE FACTORS INFLUENCING CANDIDATE DROPOUT IN RECRUITMENT PROCESS
Anandavathini K, Dr. Raja. S
DOI: 10.17148/IJARCCE.2026.154203
Abstract: Candidate drop-out is a continuing problem for organizations wishing to attract and retain top talent within the recruitment process. The purpose of this study was to identify the variety of factors contributing to candidates withdrawing or disengaging from the hiring cycle process prior to the end. The key prescriptive factors identified within this study can be categorized into the following categories: lengthy and complex recruitment processes; poor or lack of effective/efficient communication; candidacy/job expectation misalignment; and poor candidate experiences.
This study also identified external elements (e.g. competing job offers, personal circumstances, and perception of the employer brand) as contributors to candidates' dropout behaviour. The results of this research highlight inefficiencies in recruiting strategies, to include ambiguous job descriptions, protracted timeframes between stages of the recruitment process, and non-transparent selection criteria, as contributors to a candidate's commitment to accepting an employment offer.
The research illustrates the importance of establishing efficient and effective recruitment and selection procedures, open lines of communication, and positive applicant candidate experience as viable means to reduce candidate dropout rates.
Keywords: Candidate Drop-out, Recruitment Process, Talent Acquisition, Candidate Experience Hiring Delays.
This study also identified external elements (e.g. competing job offers, personal circumstances, and perception of the employer brand) as contributors to candidates' dropout behaviour. The results of this research highlight inefficiencies in recruiting strategies, to include ambiguous job descriptions, protracted timeframes between stages of the recruitment process, and non-transparent selection criteria, as contributors to a candidate's commitment to accepting an employment offer.
The research illustrates the importance of establishing efficient and effective recruitment and selection procedures, open lines of communication, and positive applicant candidate experience as viable means to reduce candidate dropout rates.
Keywords: Candidate Drop-out, Recruitment Process, Talent Acquisition, Candidate Experience Hiring Delays.
Abstract
Role of Social Media in Creating Brand Awareness among Rural Farmers
Aishwarya. S, Dr. S. Arul Krishnan
DOI: 10.17148/IJARCCE.2026.154204
Abstract: Social media is being recognized in today's "digital age" as one of the world's most inexpensive and powerful communication and marketing tools available to all types of businesses. Companies in all industries have been able to increase their ability to reach customers from remote, rural areas with the improved availability of both internet access and mobile phones. In the agricultural industry, for example, many farmers are using multiple forms of digital technology to access information about many different aspects of agriculture including how to farm, the types of agricultural inputs available, and market information. Farmers are using popular social media platforms (e.g., Facebook, YouTube, WhatsApp, and Instagram) to quickly obtain information about seeds, fertilizers, pesticides, and equipment for farming purpose. By using different forms of social media marketing, agricultural companies are able to place targeted digital ads, share videos, provide product demonstrations, and provide general informational content to farmers. The purpose of this study is to determine the extent to which social media creates awareness of brands among rural farmers. The focus of the study will assess how social media has been able to enhance rural farmers' knowledge and awareness of agricultural brands, knowledge of benefits associated with products sold by these agricultural brands, and how these factors have influenced the purchase decisions made by rural farmers. According to research conducted in this area, social media has an important role to play when it comes to communicating information about agriculture products and brands to farmers living in rural locations. In addition, findings support that farmers are able to better understand advances in agricultural technology through the use of digital platforms and to compare products from multiple producers. Consequently, rural farmers will be more likely to make educated decisions regarding their agricultural purchases. Nonetheless, there are numerous barriers that still impede rural farmers' adoption of social media, such as low levels of digital literacy and inadequate internet access.
Abstract
ScholarGrid: A Full-Stack Academic Collaboration Platform with Community-Driven Quality Layers
Ranvir Wadhawan, Swaroop Mandal, Aryan Yadav, and Meet Patil, Prof. K.L. More
DOI: 10.17148/IJARCCE.2026.154205
Abstract: Academic collaboration in higher education is frequently hindered by fragmented resources across unorganized platforms. This paper introduces ScholarGrid, a centralized full-stack solution designed to streamline note- sharing and academic communication. Built using React 19 and FastAPI, the platform implements a community-driven quality layer featuring star ratings and peer reviews to ensure content reliability. Additionally, the integration of Socket.io enables real-time, subject-specific chat channels. Our methodology employs an iterative development model, resulting in a scalable architecture that bridges the gap between informal messaging and structured learning management systems.
Keywords: Note-Sharing Platform, Full-Stack Development, Academic Collaboration, React 19, FastAPI, Peer-Review Systems.
Keywords: Note-Sharing Platform, Full-Stack Development, Academic Collaboration, React 19, FastAPI, Peer-Review Systems.
Abstract
“Next Generation Learning: Augmented Reality in Smart Education System”
Shivam, Priyanshu Ahlawat, Riya Sharma, Tanu Gupta, Rahul Vashishth, Miss. Taniya Jain, Dr. Uruj Jaleel, Dr. Satish Kumar Soni
DOI: 10.17148/IJARCCE.2026.154206
Abstract: Technology in education can influence students to learn actively and can motivate them, leading to an effective process of learning. Previous research has identified the problem that technology will create a passive learning process if the technology used does not promote critical thinking, meaning-making or metacognition. Since its introduction, augmented reality (AR) has been shown to have good potential in making the learning process more active, effective and meaningful. This is because its advanced technology enables users to interact with virtual and real-time applications and brings the natural experiences to the user. In addition, the merging of AR with education has recently attracted research attention because of its ability to allow students to be immersed in realistic experiences. Therefore, this concept paper reviews the research that has been conducted on AR. The review describes the application of AR in a number of fields of learning including Medicine, Chemistry, Mathematics, Physics, Geography, Biology, Astronomy and History. This paper also discusses the advantages of AR compared to traditional technology (such as e-learning and courseware) and traditional teaching methods (chalk and talk and traditional books). The review of the results of the research shows that, overall, AR technologies have a positive potential and advantages that can be adapted in education. The review also indicates the limitations of AR which could be addressed in future research.
Augmented reality (AR), a cutting-edge technology, has the potential to change the way students learn by superimposing virtual items and information onto the real environment. Through more immersive and interesting interactions with digital content, AR might help students better understand difficult concepts and boost their drive to learn. As a result of its contribution to student learning, AR has become increasingly appealing to educational researchers. This study aimed to descriptively explore the characteristics of AR studies in education and to qualitatively analyze the most influential ones indexed in the Web of Science (WoS) between 2000 and 2022. A scoping review was conducted to determine the sample of the AR studies in education based on the inclusion criteria. Accordingly, descriptive analyses were conducted to identify the characteristics of the AR studies in education between 2000 and 2022 in terms of publication year, country, affiliations, journals, funding agencies, and citation trends Then, the research methodologies and implications were found among the most influential AR studies in education between 2000 and 2022 by synthesizing qualitatively. The overall results indicated that AR studies in education have been conducted since 2008, with an increasing number of studies over time. Based on the implications of the most influential studies identified in terms of citation numbers, it was detected that AR has the potential to enhance education and training by providing interactive and engaging environments, linking real-world contexts with digital resources, and promoting efficiency and effectiveness in learning. [This paper was published in: "EJER Congress 2023 International Eurasian Educational Research Congress Conference Proceedings," Ani Publishing, 2023, pp. 273-291]
Augmented Reality has a positive impact in the world of education. Systematic review studies and article reviews have provided evidence of this positive impact. Augmented reality is an effective medium for learning, especially during a ISSN (O) 2278-1021, ISSN (P) 2319-5940IJARCCE
International Journal of Advanced Research in Computer and Communication Engineering Impact Factor 8.471Peer-reviewed & Refereed journalVol. 15, Issue 4, April 2026 DOI: 10.17148/IJARCCE.2026.154206 pandemic. The keyword used to find the articles on the Scopus database was “augmented AND reality AND in AND science AND education,” which discovered 1153 publications, then screening found 68 articles that met the criteria for analysis were obtained by following the Preferred Reporting Items for Systematic Review procedure. The bibliometric finding in this study is the increasing trend of augmented reality research. Research trends consist of distribution year, research type, keywords, author nationality, and international collaboration. The findings reveal that there has been an increase in research from 2018-2023 on the theme of augmented reality. The research employed several methods, i.e., qualitative, quantitative, mixed-methods, and survey. The author publishes the theme of augmented reality from 28 countries, especially the Malaysia, China, Romania, Spain, Turkey. Furthermore, it was found that the number of collaborating publications (universities or countries) was higher than those not collaborating. The AR trend in education has benefits for students and teachers. The conclusions from the analysis of the article show that the theme of augmented reality is becoming a trend in education.
The study of augmented reality, also known as AR, is currently a high priority for multiple prestigious international universities and research institutions. This research paper explores the usage of augmented reality in the world of education. It has often been found that when presented visually, information is better understood by students in the classroom. AR is such a technology that can be used to provide interactive learning visual experience giving learners a sense of deeper immersion and significantly raising the learner’s appeal and interest. Augmented reality (AR) gaming can be utilized as an instructional tool, in addition to visual learning through animations and 3D models as it has been found a useful technique for improving understanding of new information. The purpose of this research is also to illustrate an educational application that uses AR and gaming to enhance conceptual understanding and adopts a different pedagogy from traditional methods. Blender and Unity3D were used to create the application that is discussed in the paper.
Keywords: Augmented Reality, Technology, Education, Student and literature review.
Augmented reality (AR), a cutting-edge technology, has the potential to change the way students learn by superimposing virtual items and information onto the real environment. Through more immersive and interesting interactions with digital content, AR might help students better understand difficult concepts and boost their drive to learn. As a result of its contribution to student learning, AR has become increasingly appealing to educational researchers. This study aimed to descriptively explore the characteristics of AR studies in education and to qualitatively analyze the most influential ones indexed in the Web of Science (WoS) between 2000 and 2022. A scoping review was conducted to determine the sample of the AR studies in education based on the inclusion criteria. Accordingly, descriptive analyses were conducted to identify the characteristics of the AR studies in education between 2000 and 2022 in terms of publication year, country, affiliations, journals, funding agencies, and citation trends Then, the research methodologies and implications were found among the most influential AR studies in education between 2000 and 2022 by synthesizing qualitatively. The overall results indicated that AR studies in education have been conducted since 2008, with an increasing number of studies over time. Based on the implications of the most influential studies identified in terms of citation numbers, it was detected that AR has the potential to enhance education and training by providing interactive and engaging environments, linking real-world contexts with digital resources, and promoting efficiency and effectiveness in learning. [This paper was published in: "EJER Congress 2023 International Eurasian Educational Research Congress Conference Proceedings," Ani Publishing, 2023, pp. 273-291]
Augmented Reality has a positive impact in the world of education. Systematic review studies and article reviews have provided evidence of this positive impact. Augmented reality is an effective medium for learning, especially during a ISSN (O) 2278-1021, ISSN (P) 2319-5940IJARCCE
International Journal of Advanced Research in Computer and Communication Engineering Impact Factor 8.471Peer-reviewed & Refereed journalVol. 15, Issue 4, April 2026 DOI: 10.17148/IJARCCE.2026.154206 pandemic. The keyword used to find the articles on the Scopus database was “augmented AND reality AND in AND science AND education,” which discovered 1153 publications, then screening found 68 articles that met the criteria for analysis were obtained by following the Preferred Reporting Items for Systematic Review procedure. The bibliometric finding in this study is the increasing trend of augmented reality research. Research trends consist of distribution year, research type, keywords, author nationality, and international collaboration. The findings reveal that there has been an increase in research from 2018-2023 on the theme of augmented reality. The research employed several methods, i.e., qualitative, quantitative, mixed-methods, and survey. The author publishes the theme of augmented reality from 28 countries, especially the Malaysia, China, Romania, Spain, Turkey. Furthermore, it was found that the number of collaborating publications (universities or countries) was higher than those not collaborating. The AR trend in education has benefits for students and teachers. The conclusions from the analysis of the article show that the theme of augmented reality is becoming a trend in education.
The study of augmented reality, also known as AR, is currently a high priority for multiple prestigious international universities and research institutions. This research paper explores the usage of augmented reality in the world of education. It has often been found that when presented visually, information is better understood by students in the classroom. AR is such a technology that can be used to provide interactive learning visual experience giving learners a sense of deeper immersion and significantly raising the learner’s appeal and interest. Augmented reality (AR) gaming can be utilized as an instructional tool, in addition to visual learning through animations and 3D models as it has been found a useful technique for improving understanding of new information. The purpose of this research is also to illustrate an educational application that uses AR and gaming to enhance conceptual understanding and adopts a different pedagogy from traditional methods. Blender and Unity3D were used to create the application that is discussed in the paper.
Keywords: Augmented Reality, Technology, Education, Student and literature review.
Abstract
AI-Based Spam Message Shield with WhatsApp Message Detection using NLP
Arti Kumari, Akansha Sharma, Ankita Katheria, Rooban Agrawal, Satish Kumar Soni, Uruj Jaleel
DOI: 10.17148/IJARCCE.2026.154207
Abstract: With the exponential growth of mobile communication, SMS spam has become one of the most prevalent security and privacy concerns. Spam messages lead to financial fraud, data theft, and reduced user experience. Detecting such messages using traditional rule-based systems has proven insufficient due to evolving spam patterns. This research paper presents a comprehensive study of SMS spam detection techniques using machine learning models such as Naïve Bayes, Support Vector Machines (SVM), Logistic Regression, and Deep Learning approaches. A mini- project implementation demonstrates the use of natural language processing (NLP) techniques, including tokenization, TF-IDF, stemming, and lemmatization. The study highlights dataset characteristics, feature engineering, model performance, comparative results, and implementation constraints. Findings show that ML-based classifiers significantly outperform rule-based systems, achieving accuracy above 95%. Future directions include hybrid deep learning models and real-time adaptive systems.
Keywords: SMS Spam Detection, Machine Learning, NLP, TF-IDF, Classification, Naïve Bayes, Logistic Regression.
Keywords: SMS Spam Detection, Machine Learning, NLP, TF-IDF, Classification, Naïve Bayes, Logistic Regression.
Abstract
The Role of Nmap in Modern Network Security
Saurav Saini, Nikhil Singhal, Shekhar, Tanya Jain, Uruj Jaleel, Satish Kumar Soni
DOI: 10.17148/IJARCCE.2026.154208
Abstract: Network assaults have been common, resulting in the theft of private data. Information gathering is the first step that hackers do before launching an attack. Nmap is one of the most often used scanning programs at this point to gather data from the target host. To help with the ensuing attack, the acquired data can be further examined. Hence, a reliable method of identifying Nmap scanning behavior must be developed. In Nmap we can scan all the 65535 ports in one go with the packet customizable option. The intrusion detection system (IDS) frequently employs the ET OPEN rule set to safeguard hosts against nefarious intrusion.[1]
Among various tools available, Nmap (Network Mapper) stands out as one of the most powerfull as widely used open- source tools for network discovery and security auditing[1][2]
This research paper explores the role of Nmap in modern network security, focusing on its functionalities, applications, and effectiveness in vulnerability assessment and penetration testing. The study also examines various scanning techniques such as TCP, UDP, and SYN scans, along with advanced features like OS detection and the Nmap Scripting Engine (NSE) [3][4]. Additionally, the paper discusses ethical considerations, legal implications, and challenges associated with network scanning.
Through analysis of existing literature and practical use cases, this research highlights how Nmap contributes significantly to enhancing cybersecurity by enabling administrators and ethical hackers to identify weaknesses in network infrastructures [2][5].
Keywords: Network Security, Nmap, Network Scanning, Cybersecurity, Vulnerability Assessment, Penetration Testing
Among various tools available, Nmap (Network Mapper) stands out as one of the most powerfull as widely used open- source tools for network discovery and security auditing[1][2]
This research paper explores the role of Nmap in modern network security, focusing on its functionalities, applications, and effectiveness in vulnerability assessment and penetration testing. The study also examines various scanning techniques such as TCP, UDP, and SYN scans, along with advanced features like OS detection and the Nmap Scripting Engine (NSE) [3][4]. Additionally, the paper discusses ethical considerations, legal implications, and challenges associated with network scanning.
Through analysis of existing literature and practical use cases, this research highlights how Nmap contributes significantly to enhancing cybersecurity by enabling administrators and ethical hackers to identify weaknesses in network infrastructures [2][5].
Keywords: Network Security, Nmap, Network Scanning, Cybersecurity, Vulnerability Assessment, Penetration Testing
Abstract
An AI-Automated Diagnosis System for Pneumonia Using Xception CNN
Aruna S, Deepika S, Harini M, Maheshwari B
DOI: 10.17148/IJARCCE.2026.154209
Abstract: Pneumonia is a severe respiratory infection that continues to be a major cause of illness and death worldwide, particularly among children, elderly individuals, and immunocompromised patients. Timely and accurate diagnosis plays a critical role in reducing complications and improving patient survival rates. Chest X-ray imaging is one of the most commonly used diagnostic tools for pneumonia detection; however, traditional diagnosis relies heavily on manual interpretation by radiologists, which is time-consuming, subjective, and prone to human error due to fatigue and variations in expertise. These challenges are further intensified in rural and resource-limited healthcare environments where experienced radiologists are often unavailable, leading to delayed or inaccurate diagnoses. With the exponential growth of medical imaging data, there is an increasing demand for automated and intelligent diagnostic systems that can assist healthcare professionals in clinical decision-making. This project presents an automated pneumonia classification system based on deep learning techniques using chest X-ray images. The proposed system employs the Xception Convolutional Neural Network (CNN), which is well-known for its ability to extract complex and discriminative features from medical images. Transfer learning is utilized by leveraging pre-trained weights to enhance model performance and reduce training time, especially when dealing with limited labeled medical datasets. The collected chest X-ray images undergo preprocessing steps to ensure data uniformity and quality before being divided into training, validation, and testing sets. The trained model classifies images into pneumonia-affected and normal categories with improved accuracy, sensitivity, and specificity compared to traditional methods. By integrating the developed model into a computer-aided diagnosis framework, the system provides consistent and reliable support to radiologists, reduces diagnostic workload, and minimizes human intervention. Overall, the proposed solution aims to improve early pneumonia detection, enhance diagnostic efficiency, and contribute to better healthcare outcomes, particularly in underserved and remote regions.
Keywords: Pneumonia detection, chest x-ray, deep learning, Xception CNN, diagnosis system
Keywords: Pneumonia detection, chest x-ray, deep learning, Xception CNN, diagnosis system
Abstract
IntelliInterview: An AI-Based Interview Training Platform Using Natural Language Processing
Aditya Kumar, Amisha Jain, Akshit Kumar, Ajit Singh, Dr. Uruj Jaleel, Dr. Satish Kumar Soni
DOI: 10.17148/IJARCCE.2026.154210
Abstract: The growing competitiveness of the modern job market demands that candidates possess not only domain knowledge but also refined communication skills, structured thinking, and interview confidence. A significant gap exists between theoretical academic preparation and practical interview performance, particularly among students and fresh graduates who lack access to structured mock interview resources. This paper presents IntelliInterview, an AI-based Interview Training Platform that leverages Natural Language Processing (NLP) and Artificial Intelligence (AI) to simulate realistic interview environments. The system acts as a virtual interviewer, presenting domain-specific and HR interview questions, accepting text or voice responses from users, and analyzing the quality of answers based on grammar, relevance, coherence, and completeness. Following analysis, the system generates constructive feedback along with a performance score, enabling candidates to self-assess and improve iteratively. Developed using Python, the platform integrates Streamlit as the user interface framework, NLP libraries for response analysis, and speech recognition libraries for voice-based input. Experimental results indicate that repeated practice sessions using IntelliInterview lead to measurable improvements in user response quality and confidence, demonstrating the practical viability of AIpowered tools for professional skill development. This work contributes to the growing body of research on intelligent tutoring systems and conversational AI applications in the domain of career development and education.
Abstract
Image Recognition Using Convolutional Neural Networks
Krish Kumar, Ujjwal Panwar, Mohit, Ajit Singh, Satish Kumar Soni, Uruj Jaleel
DOI: 10.17148/IJARCCE.2026.154211
Abstract: Convolutional Neural Networks (CNNs) have emerged as one of the most powerful and widely adopted deep learning architectures for image recognition tasks. This paper presents a comprehensive study of CNN-based image recognition systems, examining their architectural components, working mechanisms, and practical applications across various domains. CNNs have demonstrated exceptional performance in benchmark datasets such as ImageNet, CIFAR-10, and MNIST, significantly outperforming traditional machine learning approaches. The study explores key CNN architectures including LeNet, AlexNet, VGGNet, ResNet, and Inception, analyzing their structural innovations and contributions to improving recognition accuracy. Furthermore, the paper addresses common challenges such as overfitting, vanishing gradients, and computational cost, along with techniques such as data augmentation, dropout, and batch normalization used to overcome these issues. Experimental results indicate that deep CNN models achieve accuracy rates exceeding 95% on standard benchmarks, demonstrating their effectiveness for real-world image classification tasks. The paper concludes by discussing future directions including lightweight models for edge deployment and the integration of attention mechanisms for improved recognition performance.
Keywords: Convolutional Neural Networks (CNN), Image Recognition, Deep Learning, AlexNet, ResNet, VGGNet, Transfer Learning, Object Detection, Computer Vision, Feature Extraction.
Keywords: Convolutional Neural Networks (CNN), Image Recognition, Deep Learning, AlexNet, ResNet, VGGNet, Transfer Learning, Object Detection, Computer Vision, Feature Extraction.
Abstract
Web-Based Explainable Credit Card Fraud Detection Using SMOTE, Ensemble Feature Selection, and XGBoost
A. Ruba, Hema Lakshmi L, Vasumathy A
DOI: 10.17148/IJARCCE.2026.154212
Abstract: The rapid growth of digital has enhanced the risk of fraud using the credit cards, resulting into the massive financial losses. This paper describes an explainable fraud detector system based on machine learning and Explainable Artificial Intelligence (XAI) as a web-based system. Synthetic Minority Oversampling Technique (SMOTE) is employed in the management of class imbalance in transaction data. A voting-based feature selection strategy that combines the importance of Random Forest with L1-regularized Logistic Regression (LASSO), and Chi-Square test is used to select the most significant features. An XGBoost classifier is trained on the selected features in order to predict fraud effectively. The system is implemented in the form of Flask web application, which allows real-time entry of transactions and uploading of CSV file. All transactions are categorized as fraudulent or normal with a probability score attached to it. Many features of SHAP are used to guarantee transparency including global and local feature importance, and LIME generates explanations on an instance level. The system presented has attained a high detection performance with interpretability and real life application which renders it appropriate in a financial fraud monitoring and decision support in the real world.
Keywords: credit card fraud detection, XGBoost, SMOTE, explainable AI, SHAP, LIME, web-based system.
Keywords: credit card fraud detection, XGBoost, SMOTE, explainable AI, SHAP, LIME, web-based system.
Abstract
BorrowBridge: A Community-Based Shared Resource Platform for Sustainable Consumption
Aparna Chavan, Tanmay Hursale, Aayush Jadhav, Prof. Shilpa Tandale
DOI: 10.17148/IJARCCE.2026.154213
Abstract: Borrow Bridge is a web-based platform designed to facilitate the borrowing and lending of items among users within a community. The system aims to reduce unnecessary purchases, promote resource sharing, and create a sustainable environment through collaborative consumption. The platform provides a centralized interface where users can list items, request to borrow, approve requests, and manage transactions efficiently.
The project focuses on real-time interaction, user-friendly design, and structured item management. Borrow Bridge integrates essential modules such as user authentication, item listing, request handling, and transaction tracking. This paper discusses the motivation, objectives, system architecture, functional modules, advantages, limitations, and future scope of the platform. The system demonstrates how modern web technologies can be utilized to solve real-world problems related to resource optimization and peer-to-peer sharing.
Keywords: Borrow Bridge, Item Sharing System, Web Application, Lending Platform, Resource Management, Peer-to- Peer System.
The project focuses on real-time interaction, user-friendly design, and structured item management. Borrow Bridge integrates essential modules such as user authentication, item listing, request handling, and transaction tracking. This paper discusses the motivation, objectives, system architecture, functional modules, advantages, limitations, and future scope of the platform. The system demonstrates how modern web technologies can be utilized to solve real-world problems related to resource optimization and peer-to-peer sharing.
Keywords: Borrow Bridge, Item Sharing System, Web Application, Lending Platform, Resource Management, Peer-to- Peer System.
Abstract
Language Agnostic Conversational Intelligence System For Smart Campus
M Chandana¹, Parimala c², R M Harshitha ³, Saniya Hundekar ⁴, Dr.Muhibur Rahman T.R5
DOI: 10.17148/IJARCCE.2026.154214
Abstract: In recent years, artificial intelligence and machine learning have increasingly been adopted in educational institutions to enhance smart campus governance and automate administrative processes. This paper presents a Multilingual Retrieval-Augmented Generation (RAG)-based Conversational Intelligence System designed to facilitate seamless interaction between students, faculty, and campus administration. The system integrates natural language processing, machine learning, and large language models to support real-time, context-aware communication across multiple languages. It enables efficient handling of campus-related queries such as academic schedules, fee management, grievance redressal, and event information through an intelligent conversational interface.
The proposed framework introduces a structured, multi-tier architecture that evolves from basic query-response systems to fully integrated intelligent campus platforms with personalization and decision-support capabilities. Performance metrics such as response accuracy, contextual relevance, latency, scalability, and multilingual adaptability are considered for evaluation. Existing systems often lack the integration of multilingual support, real-time data retrieval, personalized responses, and conversational intelligence within a unified framework. This paper identifies these gaps and outlines future directions for developing scalable, inclusive, and intelligent smart campus ecosystems using RAG-based approaches.
The proposed framework introduces a structured, multi-tier architecture that evolves from basic query-response systems to fully integrated intelligent campus platforms with personalization and decision-support capabilities. Performance metrics such as response accuracy, contextual relevance, latency, scalability, and multilingual adaptability are considered for evaluation. Existing systems often lack the integration of multilingual support, real-time data retrieval, personalized responses, and conversational intelligence within a unified framework. This paper identifies these gaps and outlines future directions for developing scalable, inclusive, and intelligent smart campus ecosystems using RAG-based approaches.
Abstract
Three Phases Fault Detection System
Dr. Rahat Ullah Khan, Himanshu Verma
DOI: 10.17148/IJARCCE.2026.154215
Abstract: Detecting faults in transmission lines is something we cannot afford to ignore if we want power systems to remain reliable and safe. In real-world conditions, transmission lines routinely face threats from lightning, insulation breakdown, equipment failure, and environmental disturbances. These events can trigger single line-to-ground faults, line-to-line faults, double line-to-ground faults, or full three-phase faults. Getting a grip on these faults quickly—before they cascade into larger failures—is what separates a robust power system from one that causes extended outages and equipment damage.
Keywords: fault detection, transmission line protection, ACS712, Arduino Uno, NodeMCU, GPS module, power system reliability.
Keywords: fault detection, transmission line protection, ACS712, Arduino Uno, NodeMCU, GPS module, power system reliability.
Abstract
REAL TIME SUSPECT DETECTION AND TRACKING USING AI BASED SURVELIIENCE SYSTEM
Jeffy John Binu C, Iniyan R G, Baskar P N, Mohammad Adhil KV A
DOI: 10.17148/IJARCCE.2026.154216
Abstract: This project presents an AI-based intelligent system that integrates Natural Language Processing (NLP) and Computer Vision techniques to automate resume analysis and real-time surveillance monitoring. The system utilizes advanced machine learning and deep learning models such as Named Entity Recognition (NER), semantic similarity algorithms, Convolutional Neural Networks (CNN), and YOLO-based object detection to process both textual and visual data efficiently. It analyzes resumes to match job descriptions, identify skill gaps, and provide optimization recommendations, while in surveillance it focuses on detecting and identifying suspicious persons based on given image data and tracking them across video frames. The system compares live video input with pre-stored images to recognize individuals and assign unique tracking identities for continuous monitoring. By combining text analysis with image- based person recognition, the system improves accuracy, reduces manual effort, and enables faster decisionmaking. Overall, the proposed system provides an efficient, scalable, and real-time solution for recruitment optimization and intelligent surveillance applications.
Keywords: Artificial Intelligence (AI), Natural Language Processing (NLP), Computer Vision (CV), Deep Learning (DL), Resume Analysis, Semantic Similarity, Named Entity Recognition (NER), Convolutional Neural Networks (CNN), YOLO (You Only Look Once).
Keywords: Artificial Intelligence (AI), Natural Language Processing (NLP), Computer Vision (CV), Deep Learning (DL), Resume Analysis, Semantic Similarity, Named Entity Recognition (NER), Convolutional Neural Networks (CNN), YOLO (You Only Look Once).
Abstract
Home Blood Sample Collection System: A Smart Health Portal for Home Diagnostics
M Tharun Kumar, Rohith Edwin, ShivKumar Gowda, Sachin E, Dr. Muhibur Rahaman T.R
DOI: 10.17148/IJARCCE.2026.154217
Abstract: The rapid advancement of digital healthcare technologies has significantly increased the demand for convenient, accessible, and efficient diagnostic services. Traditional blood testing methods require patients to physically visit diagnostic laboratories, leading to time consumption, inconvenience, overcrowding, and limited accessibility, especially for elderly individuals, working professionals, and patients in rural areas. This paper presents a Home Blood Sample Collection System, a smart health portal designed to streamline the process of booking blood tests, scheduling home sample collection, and delivering reports digitally. The proposed system integrates patients, laboratories, administrators, and phlebotomists through a centralized web-based platform, enabling seamless communication and efficient workflow management. It provides features such as online test selection, appointment scheduling, real-time tracking of sample collection, and secure digital report access, thereby eliminating manual processes and reducing human errors. The system ensures data privacy and reliability while improving transparency and user experience in diagnostic services. By reducing the need for hospital visits and enabling doorstep healthcare services, the platform enhances accessibility and operational efficiency. Furthermore, the system is scalable and can be extended with additional features such as online payment integration, teleconsultation, and advanced health analytics, making it a promising solution for modern digital healthcare ecosystems.
Keywords: Home Blood Sample Collection, Home Diagnostics, Healthcare System, Web Application, Telemedicine, Digital Health, Blood Test Booking, Report Management
Keywords: Home Blood Sample Collection, Home Diagnostics, Healthcare System, Web Application, Telemedicine, Digital Health, Blood Test Booking, Report Management
Abstract
AI Driven Railway Track Crack Detection And Classification With YOLOV
Koppisetti Sriram, Gonuguntla Brunda, K. Akila
DOI: 10.17148/IJARCCE.2026.154218
Abstract: The Railway Track Crack Detection system is an intelligent deep learning–based solution designed to automatically detect cracks and defects in railway tracks using image data. The system utilizes advanced computer vision techniques and the YOLOv5 object detection model to identify crack regions with high accuracy and speed. It integrates data collection, preprocessing, image annotation using tools like LabelImg, model training, testing, and real-time detection into a unified workflow. The trained model detects cracks in input images by generating bounding boxes along with confidence scores, enabling clear visualization of defects. The system also includes performance evaluation, parameter tuning, and optimization to improve accuracy and reduce false detections. In addition, it supports real-time monitoring through continuous analysis of images or video frames, making it suitable for practical railway inspection scenarios. Compared to traditional manual inspection methods, this approach reduces human effort, minimizes errors, and enables early detection of potential failures. By leveraging deep learning and object detection techniques, the project provides a cost-effective, scalable, and efficient solution for improving railway safety and maintenance.
Keywords: Railway Track Crack Detection, YOLOv5, Deep Learning, Computer Vision, Object Detection, Image Processing, Defect Detection, Real-Time Monitoring, Railway Safety, Predictive Maintenance
Keywords: Railway Track Crack Detection, YOLOv5, Deep Learning, Computer Vision, Object Detection, Image Processing, Defect Detection, Real-Time Monitoring, Railway Safety, Predictive Maintenance
Abstract
Sports Management System
Nitin Chaudhari, Mayur Gaikwad, Girish Kale, Amit Kanojiya, Prof. Madhuri Dalal
DOI: 10.17148/IJARCCE.2026.154219
Abstract: Managing sports programs in educational institutions and clubs has always involved a fair amount of juggling — player rosters, match schedules, tournament brackets, and performance records all demanding attention at once. This paper presents a Sports Management System designed to bring all of these responsibilities under one digital roof. Built around a centralized database and a role-based interface, the system lets administrators, coaches, and players each interact with the information most relevant to them, without stepping on each other’s toes. The result is a platform that cuts down on paperwork, reduces scheduling conflicts, and makes it significantly easier to track how individual athletes and teams are performing over time. We describe the motivation behind the system, review related work in the field, outline the proposed architecture and key features, and discuss the practical benefits the system offers to real-world sports programs.
Keywords: Sports Management, Player Management, Team Management, Match Scheduling, Tournament Organization, Score Tracking, Database Management, Performance Analytics
Keywords: Sports Management, Player Management, Team Management, Match Scheduling, Tournament Organization, Score Tracking, Database Management, Performance Analytics
Abstract
Crowd Abnormal Behaviour Detection Using YOLO V
Punitha N, Vasu Devan B, Sundhara Chozhan K, Tamil Selvan B, Thavasi M
DOI: 10.17148/IJARCCE.2026.154220
Abstract: is paper presents a Crowd Abnormal Behavior Detection system using advanced deep learning techniques for real-time surveillance and safety monitoring. The system utilizes the YOLO (You Only Look Once) object detection algorithm to identify human activities in crowded environments and classify them as normal or abnormal. Abnormal behaviors such as fighting, running, panic situations, and sudden crowd movements are detected with high accuracy. The proposed system processes video streams in real-time and provides quick alerts, helping authorities take immediate action. The integration of computer vision and artificial intelligence enhances public safety in areas such as railway stations, shopping malls, and public gatherings.
Keywords: Crowd Analysis, Abnormal Behavior Detection, YOLO, Deep Learning, Computer Vision
Keywords: Crowd Analysis, Abnormal Behavior Detection, YOLO, Deep Learning, Computer Vision
Abstract
ATS Resume NLP Analyzer: A Hybrid, Explainable, and Practical Framework for Resume-Job Matching
MD Auranzeb Khan, Srijan Mani Tripathi, Kunal, Durga Devi
DOI: 10.17148/IJARCCE.2026.154221
Abstract: This paper proposes and experimentally validates a production-ready hybrid framework for automated resume–job matching using explainable Natural Language Processing (NLP) techniques. The system processes heterogeneous resume formats and job descriptions to compute an interpretable Applicant Tracking System (ATS) score by integrating deterministic skill matching, semantic similarity using sentence embeddings, and lexical keyword overlap.
We construct a modular pipeline that transforms unstructured resume text into structured representations, enabling robust comparison against job requirements. The system employs a hybrid scoring mechanism combining three signals: skill coverage, embedding-based semantic similarity, and token-level keyword matching. Additionally, the framework includes OCR-based ingestion for scanned resumes, explainable score decomposition, missing-skill diagnostics, and optional Large Language Model (LLM)-assisted feedback generation.
Experimental evaluation is conducted on a curated dataset of resumes and job descriptions with human -labeled relevance scores. The proposed system achieves strong alignment with human judgment, demonstrating improved ranking consistency over baseline keyword-only approaches. The system maintains low latency suitable for real-time deployment and includes robust fallback mechanisms for production reliability.
This work demonstrates that a hybrid deterministic-semantic approach can significantly improve transparency, usability, and effectiveness in automated recruitment systems while remaining scalable and deployable in real-world environments.
Keywords: Applicant Tracking System, Resume Screening, NLP, Semantic Similarity, Explainable AI, Recruitment Analytics, FastAPI
We construct a modular pipeline that transforms unstructured resume text into structured representations, enabling robust comparison against job requirements. The system employs a hybrid scoring mechanism combining three signals: skill coverage, embedding-based semantic similarity, and token-level keyword matching. Additionally, the framework includes OCR-based ingestion for scanned resumes, explainable score decomposition, missing-skill diagnostics, and optional Large Language Model (LLM)-assisted feedback generation.
Experimental evaluation is conducted on a curated dataset of resumes and job descriptions with human -labeled relevance scores. The proposed system achieves strong alignment with human judgment, demonstrating improved ranking consistency over baseline keyword-only approaches. The system maintains low latency suitable for real-time deployment and includes robust fallback mechanisms for production reliability.
This work demonstrates that a hybrid deterministic-semantic approach can significantly improve transparency, usability, and effectiveness in automated recruitment systems while remaining scalable and deployable in real-world environments.
Keywords: Applicant Tracking System, Resume Screening, NLP, Semantic Similarity, Explainable AI, Recruitment Analytics, FastAPI
Abstract
An intelligent immigration and safety awareness platform
Aditya Raman, Vyom Pandey, Mohammad Rayyan Basha
DOI: 10.17148/IJARCCE.2026.154222
Abstract: ImmigrationIQ is an online tool designed to assist people navigating the full scope of their immigration preparations. In total, there are five major components included in the overall structure of the tool: 1) Migration Planning System (MPS); 2) Destination Intelligence Browser; 3) Legal & Cultural Learning Hub; 4) Safety & Anti-Fraud Center; and, 5) User Dashboard that will be continually updated. The front end of the platform was built using HTML5/CSS3/Vanilla JavaScript. The back-end functionality of the platform has been implemented using Firebase Authentication and Cloud Firestore for user identity management and storing user plans on the cloud. Using Firebase as the primary option for data storage and localStorage as the secondary option, a two path approach provides continuous operation regardless of network conditions. The platform contains many other unique tools such as a First 72 hours decision simulator, a multi- step quiz engine, and real time scam alerts. Collectively, they address the fragmented and untrustworthy status of today’s general purpose immigration information platforms. The experimental testing performed shows quick loading times for all modules of less than 300 ms, low-latency communication from client to server (<1 second), and good fallback behavior when services are unavailable. Overall, this study demonstrates that developing a modular, user-centric, frontend design of an immigration preparation tool is possible to develop quickly and easily to use by both users and developers alike and that it could significantly enhance the outcomes of preparing for immigration.
Keywords: Immigration Planning, Web Application, Firebase, Semantic Web, Safety Preparedness, Decision Simulator, Modular Architecture, Personalized Roadmap, Explainable UI
Keywords: Immigration Planning, Web Application, Firebase, Semantic Web, Safety Preparedness, Decision Simulator, Modular Architecture, Personalized Roadmap, Explainable UI
Abstract
A Context-Aware Security Layer for Preventing Account Takeover Attacks in Online Transactions
N.punitha, ME., Omeshwar K, Poovarasan K, Sampath S, Shahul Hameed N
DOI: 10.17148/IJARCCE.2026.154223
Abstract: The rapid growth of digital payment systems and e-commerce platforms has significantly improved convenience but has also led to an increase in online payment frauds and account takeover attacks. Traditional One-Time Password (OTP)-based authentication systems are vulnerable to social engineering attacks such as phishing, fake customer support calls, and deceptive messages. This paper proposes a contextaware security framework that enhances OTP-based authentication by integrating transaction details, location analysis, and real-time cybercrime reporting. The system links OTPs with billing information such as merchant name, transaction amount, and masked card details, enabling users to verify transactions before authorization. Additionally, the system detects suspicious activities using location- based anomaly detection and provides instant fraud reporting with transaction blocking. This approach transforms security from a reactive to a proactive model, reducing financial losses and improving user trust in digital payment systems.
Keywords: OTP Security, Payment Fraud Detection, Context-Aware Authentication, Cybercrime Reporting, Ecommerce Security
Keywords: OTP Security, Payment Fraud Detection, Context-Aware Authentication, Cybercrime Reporting, Ecommerce Security
Abstract
A Machine Learning-Based Precision Agriculture System for Crop Recommendation and Yield Prediction
Alok Kumar Dwivedi, Harshit Tiwari, Yogendra Kumar, Kumar Bibhuti Bhushan Singh
DOI: 10.17148/IJARCCE.2026.154224
Abstract: Agriculture plays a vital role in the economic and social development of India, yet many farmers continue to rely on traditional decision-making practices for crop selection and production planning. Such conventional methods often lead to inappropriate crop choices, inefficient resource utilization, and inconsistent yield outcomes. To address these challenges, this paper presents a machine learning-based precision agriculture system for crop recommendation and yield prediction. The proposed system utilizes a Random Forest Classifier to recommend the most suitable crop based on critical soil and environmental parameters such as nitrogen, phosphorus, potassium, temperature, humidity, pH, and rainfall. In addition, an ensemble regression framework combining Random Forest and XGBoost is employed to predict crop yield using historical agricultural and climatic data from different Indian states. The system also integrates real-time weather support and expert consultation features to improve decision-making for farmers. Experimental results demonstrate a crop recommendation accuracy of 96.3% and an R² score of 0.887 for yield prediction, indicating high reliability and practical applicability. The proposed framework provides an intelligent, scalable, and farmer-centric solution for smart agriculture and precision farming applications, with significant potential to improve agricultural productivity and support data-driven farming decisions.
Abstract
Edu-Vision: An AI-Powered Multimodal Educational Assistant for Intelligent Content Understanding and Personalized Tutoring
Sura Reddy, George A, Abhishek M, Dr.Paavai Anand
DOI: 10.17148/IJARCCE.2026.154225
Abstract: Edu-Vision is an advanced AI-powered educational assistant designed to understand, interpret, and teach educational content from multiple formats including PDF files, Word documents, PowerPoint presentations, images, diagrams, and scanned notes. The system integrates large language models, optical character recognition, vision-language models, and real-time internet retrieval to deliver explanations, summaries, quizzes, flashcards, study plans, and personalized tutoring support. The implementation combines file-specific extraction modules, tutoring pipelines powered by language models, and diagram analysis using a vision encoder-decoder model. The platform also includes context- aware document indexing, page-by-page explanation, accessibility-oriented features such as braille-ready output and audio-friendly quizzes, and adaptive tutoring based on subject and learner needs. By following Universal Design for Learning principles, Edu-Vision improves inclusivity and supports flexible, student-centered learning. The project demonstrates how multimodal AI can transform educational assistance into an interactive, accessible, and personalized learning experience.
Keywords: Multimodal Learning, Educational Assistant, OCR, Large Language Models, Vision-Language Model, Personalized Tutoring, Universal Design for Learning.
Keywords: Multimodal Learning, Educational Assistant, OCR, Large Language Models, Vision-Language Model, Personalized Tutoring, Universal Design for Learning.
Abstract
IoT-Based Smart System for Arthritis Pain Monitoring and Relief
Abinaya S, Bavadharani R, Gajalakshmi P, Katherine R
DOI: 10.17148/IJARCCE.2026.154226
Abstract: Arthritis and knee joint disorders are among the most prevalent musculoskeletal conditions globally, causing chronic pain, reduced mobility, and difficulty in daily activities. Conventional healthcare systems rely on periodic hospital visits, subjective pain assessments, and manual therapies, which fail to provide continuous monitoring or timely automated relief. This paper presents an IoT-Based Smart System for Arthritis Pain Monitoring and Relief that integrates wearable sensors, embedded processing, machine learning, and IoT cloud connectivity into a unified rehabilitation solution. The system employs an IMU accelerometer (MPU6050), temperature sensor (LM35), force sensor (FSR), and heart rate sensor (MAX30100) to continuously collect real-time physiological and motion data from the patient’s knee. An Arduino UNO microcontroller processes the sensor inputs, displays readings on a 16×2 LCD, and transmits data to the cloud via an ESP8266 Wi-Fi module. A Python-based Decision Tree classifier, trained on multi-parameter sensor data, predicts pain severity as Low, Medium, or High. Based on the predicted level, the system automatically activates a Peltier module to deliver heat therapy (medium pain) or cold therapy (high pain), while recommending exercise for low pain conditions. Experimental results demonstrate accurate pain classification and real-time therapy activation, offering a cost-effective, user-friendly, and intelligent solution for home-based arthritis rehabilitation.
Keywords: IoT, Arthritis Pain Monitoring, Machine Learning, Decision Tree, Peltier Module, Wearable Sensors, Smart Healthcare, Real-Time Therapy, Arduino, Knee Rehabilitation.
Keywords: IoT, Arthritis Pain Monitoring, Machine Learning, Decision Tree, Peltier Module, Wearable Sensors, Smart Healthcare, Real-Time Therapy, Arduino, Knee Rehabilitation.
Abstract
DEEP LEARNING FRAMEWORK FOR SUPER ENHANCER PREDICTION USING CNN AND MULTI-HEAD ATTENTION WITH CROSS-SPECIES TRANSFER LEARNING
Parimala M, Naveen A, Veeradinesh R, Vikram M, Vinoth G
DOI: 10.17148/IJARCCE.2026.154227
Abstract: This paper presents a Super Enhancer Prediction system using advanced deep learning techniques for genomic analysis. The system utilizes a hybrid CNN–Transformer architecture to analyze DNA sequences and classify them as Super Enhancers (SE) or Typical Enhancers (TE). Convolutional layers capture local sequence motifs, while Multi-Head Self-Attention models long-range dependencies within genomic sequences. The proposed system incorporates cross- species transfer learning by pretraining on combined human and mouse datasets and fine-tuning for species-specific prediction. The model achieves high accuracy with improved AUC scores compared to existing methods. In addition to classification, the system provides biological insights such as GC content, motif density, and important regulatory regions. The integration of an interactive web interface allows users to upload DNA sequences and visualize results efficiently. This approach enhances genomic research by providing a fast, scalable, and interpretable solution for super enhancer identification.
Keywords: Super Enhancer Prediction, Deep Learning, CNN, Transformer, DNA Sequence Analysis, Transfer Learning, Bioinformatics, Genomics
Keywords: Super Enhancer Prediction, Deep Learning, CNN, Transformer, DNA Sequence Analysis, Transfer Learning, Bioinformatics, Genomics
Abstract
VIRTUAL TRY-ON FOR CLOTHES USING OPENCV
Bavip S, Dhanush S, Gokul Kumar S, Mokesh Raj V, Ms. MADHULIKA MK., M.E., Dept of Artificial Intelligence and Data Science
DOI: 10.17148/IJARCCE.2026.154228
Abstract: Virtual try-on systems are emerging as a transformative solution in the e-commerce domain by enabling users to visualize apparel digitally before purchase. This paper presents a lightweight and real-time virtual try-on framework developed using OpenCV and MediaPipe for accurate garment overlay. The system captures live video input through a webcam, detects human body landmarks, and overlays garments using affine transformations and alpha blending techniques. Unlike deep learning-based methods requiring extensive computational resources, the proposed system leverages classical computer vision algorithms to achieve real-time performance of 20–30 FPS on standard hardware. The system includes modules such as human detection, pose estimation, garment segmentation, overlay warping, and gesture-based interaction. Experimental evaluation shows high alignment accuracy and improved user engagement, with approximately 85% user satisfaction. The system also contributes to sustainability by reducing return rates in online shopping. Limitations include handling complex poses and dynamic fabric deformation. Future work focuses on integrating deep learning-based fitting models and 3D simulation for enhanced realism.
Keywords: Virtual Try-On, OpenCV, Computer Vision, Pose Estimation, Image Processing, Augmented Reality, E- commerce, MediaPipe
Keywords: Virtual Try-On, OpenCV, Computer Vision, Pose Estimation, Image Processing, Augmented Reality, E- commerce, MediaPipe
Abstract
Multilingual Summarization of Youtube Video using NLP
Ms.KaviPriya M.Tech, Bharath S, Jayaganesh S., Kishor S, Mithun S
DOI: 10.17148/IJARCCE.2026.154229
Abstract: With the exponential growth of digital video content on platforms like YouTube, users face significant "information overload," particularly when accessing content in foreign languages. This project proposes an automated, Multilingual YouTube Video Summarization system designed to bridge the gap between high-volume video data and efficient information consumption.
The system implements a multi-stage pipeline beginning with a Transformer-based Automatic Speech Recognition (ASR) module to transcribe audio with high robustness to noise and linguistic variations. Following transcription, the text undergoes a rigorous preprocessing phase—including tokenization and stop-word removal—before being converted into high-dimensional vector representations using Sentence-BERT (SBERT). This semantic embedding layer ensures that the core meaning of the video is preserved regardless of the source language.
The heart of the project is the Salgueiro Framework for Temporal Semantic Mapping, which facilitates Abstractive Synthesis. Unlike traditional extractive methods that merely copy sentences, our system generates a new, coherent narrative that maintains the chronological integrity of the original video.
Keywords: Multilingual NLP, YouTube Summarization, Text Summarization, Machine Translation, Transformer Models, Speech-to-Text.
The system implements a multi-stage pipeline beginning with a Transformer-based Automatic Speech Recognition (ASR) module to transcribe audio with high robustness to noise and linguistic variations. Following transcription, the text undergoes a rigorous preprocessing phase—including tokenization and stop-word removal—before being converted into high-dimensional vector representations using Sentence-BERT (SBERT). This semantic embedding layer ensures that the core meaning of the video is preserved regardless of the source language.
The heart of the project is the Salgueiro Framework for Temporal Semantic Mapping, which facilitates Abstractive Synthesis. Unlike traditional extractive methods that merely copy sentences, our system generates a new, coherent narrative that maintains the chronological integrity of the original video.
Keywords: Multilingual NLP, YouTube Summarization, Text Summarization, Machine Translation, Transformer Models, Speech-to-Text.
Abstract
BUILDING MANAGEMENT SYSTEM
Vatsal Jain, Darshan Kandalgaonkar, Gautam Enjam, Deepesh Ahire, Rakesh Suryawanshi
DOI: 10.17148/IJARCCE.2026.154230
Abstract: Managing residential buildings usually relies on standard First-In-First-Out (FIFO) ticketing. This rigid approach often causes problems because minor tasks can easily delay urgent emergencies. This paper presents a Smart Building Management System that solves this scheduling flaw by combining a dynamic triage algorithm with a resident- focused micro-economy. Instead of treating all complaints equally, the system applies an exponential aging formula to tickets, strictly capped at a maximum score of 100. The queueing engine constantly adjusts these priorities by scanning text descriptions for danger keywords, tracking SLA deadlines, and boosting scores when multiple neighbors report the exact same problem. Concurrently, a point-based gamified economy runs in the background. Automated tasks track maintenance payments and adjust resident trust scores based on their financial reliability. Our testing shows that linking contextual math to a gamified ledger successfully creates a self-correcting environment. Emergencies receive immediate action, routine tasks avoid starvation, and residents are financially incentivized to maintain good standing.
Keywords: Algorithmic Starvation, Building Management Systems, Context-Aware Triage, Gamification, Micro- Economy, Queuing Theory
Keywords: Algorithmic Starvation, Building Management Systems, Context-Aware Triage, Gamification, Micro- Economy, Queuing Theory
Abstract
Club Cycle: A Secure, Multi-Tiered Ecosystem for Campus Organization Management
Sharavana Ragav S, Aryan Sharma, Vibhu Sahu, Rajasekar Velswamy
DOI: 10.17148/IJARCCE.2026.154231
Abstract: Traditional academic environments often face significant hurdles in managing student organizations due to fragmented manual systems and inefficient communication channels, resulting in data redundancy and administrative delays. This research introduces Club-Cycle, an integrated management ecosystem designed to streamline club operations through a centralized digital interface. The architecture utilizes a Spring Boot backend and a React-based frontend, employing JSON Web Tokens (JWT) for secure authentication and Docker for containerized deployment. Key functionalities include automated membership tracking, event orchestration, role-based access control, and comprehensive analytical reporting. Performance evaluations indicate that the system maintains an average response time of 120 milliseconds under high traffic conditions and reduces manual administrative tasks by approximately 45%. Additionally, the automated validation of member credentials ensures high data integrity and security. In conclusion, Club-Cycle provides a scalable and robust framework that optimizes extracurricular administration, fostering a more organized and engaged campus community while significantly reducing operational overhead for both students and faculty members.
Keywords: Spring Boot, React.js, JSON Web Token (JWT), Role-Based Access Control (RBAC), Campus Organization Management, RESTful Web Services, Docker Containerization, Event Orchestration, Scalable Architecture.
Keywords: Spring Boot, React.js, JSON Web Token (JWT), Role-Based Access Control (RBAC), Campus Organization Management, RESTful Web Services, Docker Containerization, Event Orchestration, Scalable Architecture.
Abstract
Tamil Nadu 2026 Assembly Election Prediction Using Machine Learning and Dravidian Social Media Sentiment Analysis
Blesson Xavier M, Chirpparasan P, Hariganesh A, Kuduminathan P, Mrs. L. Shakira Banu, M.E
DOI: 10.17148/IJARCCE.2026.154232
Abstract: Virtual election forecasting using machine learning represents a transformative approach in the domain of political data science. Predicting election outcomes with accuracy has always been a challenging problem due to the complex interplay of historical voting patterns, incumbency effects, party momentum, and evolving public sentiment. This paper presents a machine learning-based framework for predicting the Tamil Nadu Legislative Assembly Election 2026 across all 234 constituencies. The system integrates Tamil Nadu Assembly Election data spanning 1971 to 2021 with real social media sentiment derived from the DravidianCodeMix dataset—comprising 43,632 Tamil code-mixed posts. A Random Forest classifier trained on seven engineered features achieved 88.84% accuracy on the 2021 validation set, outperforming the Gradient Boosting baseline of 87.12%. Sentiment analysis using a LaBSE+SVM model (NAACL DravidaLangTech 2025) was applied to 2,560 party-matched tweets to produce a sentiment score per party. The final 2026 prediction assigns 133 seats to Dravida Munnetra Kazhagam (DMK)—crossing the 118-seat majority threshold—65 to All India Anna Dravida Munnetra Kazhagam (AIADMK), and 18 to Indian National Congress (INC), consistent with the 2021 electoral outcome trend.
Keywords: Election Prediction, Machine Learning, Random Forest, Sentiment Analysis, DravidianCodeMix, Tamil Nadu 2026, LaBSE, Gradient Boosting.
Keywords: Election Prediction, Machine Learning, Random Forest, Sentiment Analysis, DravidianCodeMix, Tamil Nadu 2026, LaBSE, Gradient Boosting.
Abstract
AI TEST AUTOMATION
Irram Fatima, Mohammad Afham, Shravan Chumble
DOI: 10.17148/IJARCCE.2026.154234
Abstract: Modern test automation workflows are often fragmented across isolated scripts, CI logs, repository tools, and manual triage processes, making it difficult for teams to prioritize unstable tests, interpret recurring failures, and maintain brittle UI selectors efficiently. This project presents a local-first intelligent test automation platform that unifies test management, execution tracking, failure analysis, self-healing support, and governed repository delivery within a single system built with FASTAPI, Next.js, and PostgreSQL. The platform combines a background execution queue with persisted test and run history, a deployed Random Forest-based prioritization model for ranking failure-prone tests, a TF- IDF plus K-Means clustering pipeline for grouping similar failure patterns, an explainable per-test insight layer, heuristic selector-repair suggestions with confidence thresholds, and GitHub delivery workflows that preserve human approval before branch or pull request creation. Using a 45,000-row CI/CD failure-log dataset packaged with the project, the prioritization pipeline outperformed its heuristic fallback in offline evaluation, while the clustering pipeline produced operationally interpretable but only weakly separated failure groups. The prototype also demonstrates practical explainability through surfaced model factors, cluster keywords, review states, and approval histories. However, the current implementation remains limited by simplified or synthetic data characteristics, heuristic healing logic, and partial reliance on simulated execution paths. Overall, the project shows that multiple intelligent QA functions can be integrated into one transparent, low-cost, and locally reproducible test operations platform.
Keywords: Test Automation, Continuous Integration, Test Case Prioritization, Machine Learning in Testing, Predictive Modeling, Unsupervised Learning, Test Case Prioritization, Failure Clustering, Local-First Architecture.
Keywords: Test Automation, Continuous Integration, Test Case Prioritization, Machine Learning in Testing, Predictive Modeling, Unsupervised Learning, Test Case Prioritization, Failure Clustering, Local-First Architecture.
Abstract
Autonomous Multi-Agent Pipeline for Biomedical Research Automation
Murugeswari K, Pradeesh S, Syed Umar Nafeez G, Vimal M, Vinny Sam Francis V
DOI: 10.17148/IJARCCE.2026.154235
Abstract: This paper presents a multi-agent system for biomedical question answering that emulates a structured research workflow using large language models (LLMs). The proposed architecture leverages a LangGraph-based pipeline in which multiple specialized agents collaboratively perform literature retrieval, hypothesis generation, experimental protocol design, and validation. Three parallel junior agents gather evidence from diverse biomedical sources, followed by a supervisor that synthesizes a unified hypothesis. Subsequent agents refine the hypothesis, design experimental protocols, and conduct peer and safety reviews through an iterative feedback loop. A principal investigator agent produces a final decision, while an evaluator module acts as an LLM-as-a-judge to assess quality, precision, recall, latency, and cost. The system integrates retrieval-augmented generation (RAG), structured JSON outputs, and retry mechanisms to ensure robustness and consistency. Experimental evaluation demonstrates that the proposed approach improves reasoning depth, factual grounding, and decision reliability compared to single-agent baselines. Additionally, the framework provides transparent cost and latency tracking, making it suitable for real-world research assistance applications.
Keywords: Multi-Agent Systems, Biomedical Question Answering, Large Language Models, Retrieval-Augmented Generation, LangGraph, Experimental Protocol Design, LLM Evaluation, AI in Healthcare.
Keywords: Multi-Agent Systems, Biomedical Question Answering, Large Language Models, Retrieval-Augmented Generation, LangGraph, Experimental Protocol Design, LLM Evaluation, AI in Healthcare.
Abstract
LARGE SCALE DATA AUDIT THROUGH BIG DATA TECHNOLOGIES
Anbarasi N, Prasanna P, Praveen Kumar M, Surendar N D, Venkadesh K
DOI: 10.17148/IJARCCE.2026.154236
Abstract: This paper presents a comprehensive Big Data analytics pipeline for YouTube trending video data using Apache Spark, TextBlob NLP, MySQL, and Streamlit Dashboard technologies. The system integrates five primary components: the YouTube Data API v3 for automated collection of up to 200 trending videos and 4,000 user comments per pipeline run; a MySQL 8.0 relational database for structured storage; Apache Spark 3.4 (PySpark) for distributed data transformation; TextBlob NLP for lexicon-based sentiment analysis deployed as Spark User Defined Functions; and a Streamlit and Plotly multi-page interactive dashboard for analytics visualization. Results from a representative pipeline run revealed that Music and Entertainment categories dominate India's trending landscape, Gaming audiences exhibit the highest engagement scores (4.12%), and 52% of comments are Neutral, 33% Positive, and 15% Negative. The system demonstrates the practical application of Big Data engineering principles to social media analytics, delivering actionable intelligence on content trends, viewer engagement patterns, and audience sentiment for the Indian YouTube market.
Keywords: YouTube Analytics, Apache Spark, PySpark, TextBlob, Sentiment Analysis, Big Data Pipeline, MySQL, Streamlit, Engagement Score, Natural Language Processing, India Trending, Data Engineering
Keywords: YouTube Analytics, Apache Spark, PySpark, TextBlob, Sentiment Analysis, Big Data Pipeline, MySQL, Streamlit, Engagement Score, Natural Language Processing, India Trending, Data Engineering
Abstract
An Adaptive Stream-Native Anomaly Detection Framework Using Hybrid Unsupervised Learning
Jeevalakhmi K, Nithish T, Prathap S, Ramkumar K, Vijay M
DOI: 10.17148/IJARCCE.2026.154237
Abstract: Real-time anomaly detection in high-velocity data streams is critical for modern distributed systems, financial transactions, and IoT environments. Traditional batch-based anomaly detection techniques fail to adapt to evolving data distributions and concept drift. This paper proposes an adaptive stream-native anomaly detection framework using hybrid unsupervised learning techniques combining Isolation Forest and Autoencoder models. The system is designed over Apache Kafka-based streaming architecture to process continuous data with minimal latency. Experimental evaluation on real-world streaming datasets demonstrates improved detection accuracy and reduced false positive rates compared to standalone models. The proposed framework enables scalable, adaptive, and efficient anomaly detection in dynamic environments.
Keywords: Stream Processing, Anomaly Detection, Hybrid Unsupervised Learning, Isolation Forest, Autoencoder, Apache Kafka
Keywords: Stream Processing, Anomaly Detection, Hybrid Unsupervised Learning, Isolation Forest, Autoencoder, Apache Kafka
Abstract
SMART EMERGENCY VEHICLE ALERT SYSTEM USING ML & IoT
Parimala M, Nijanthan M, Shiam Alex S, Sibivarma S, Sugumaran
DOI: 10.17148/IJARCCE.2026.154238
Abstract: This paper presents a Smart Emergency Vehicle Alert System using Machine Learning (ML) and Internet of Things (IoT) technologies to improve road safety and reduce response time for emergency vehicles such as ambulances, fire trucks, and police vehicles. The system uses IoT sensors and GPS modules to detect the real-time location of emergency vehicles and communicates with nearby vehicles and traffic signals. Machine Learning algorithms analyze traffic patterns and predict optimal routes, while alerting nearby drivers through mobile applications or in-vehicle systems. The proposed system reduces traffic congestion, prevents accidents, and ensures faster emergency response. It provides a scalable, real-time, and intelligent solution for smart city traffic management.
Keywords: Emergency Vehicle Detection, IoT, Machine Learning, Smart Traffic System, GPS Tracking, Road Safety
Keywords: Emergency Vehicle Detection, IoT, Machine Learning, Smart Traffic System, GPS Tracking, Road Safety
Abstract
CYBERBULLYING DETECTION SYSTEM USING MACHINE LEARNING
Hovarthan S, Kishohar S, Mohamed Hathil M, Mugunthan R
DOI: 10.17148/IJARCCE.2026.154239
Abstract: Cyberbullying has become a serious issue on social media platforms, affecting individuals emotionally and psychologically through harmful messages, images, and audio content. Traditional systems for detecting cyberbullying are limited as they mainly focus on keyword-based text analysis and fail to understand context or handle multimedia data effectively. To overcome these limitations, this project proposes an intelligent cyberbullying detection system using advanced machine learning and deep learning techniques. Overall, the proposed system improves the accuracy, efficiency, and scalability of cyberbullying detection by combining multimodal data analysis and modern AI techniques, making it a practical solution for enhancing online safety and digital well-being.
Keywords: Cyberbullying Detection, Machine Learning, Natural Language Processing, BERT, Convolutional Neural Network, Image Processing, Speech Recognition, Multimodal Analysis, Social Media Analysis, Real-Time Detection
Keywords: Cyberbullying Detection, Machine Learning, Natural Language Processing, BERT, Convolutional Neural Network, Image Processing, Speech Recognition, Multimodal Analysis, Social Media Analysis, Real-Time Detection
Abstract
The Evolution and Growth of IT Sector in India
Aditi Chaudhary, Anshul Sharma, Chitwan Ahlawat, Abhay Dhiman, Abhay Verma, Usha Kumari
DOI: 10.17148/IJARCCE.2026.154240
Abstract: The paper explains the rise and growth of India’s IT industry. In general, information technology covers all aspects of managing and processing information. The last decade of 20th century has witnessed information technology to have revolutionary effect on the lives of people. In the last two decades there is 20 times increase in export revenues for the IT industry, employing over two million people. Today the whole IT industry is playing a major role in the growth of Indian economy. The paper shows how to analyse the growth and performance of information technology industry in India. Various aspects of information technology industry like composition, revenue, exports, wealth creation, size and share, localization etc. are studied.
Keywords: India, Information technology, services, Indian economy.
Keywords: India, Information technology, services, Indian economy.
Abstract
AI-POWERED VIRTUAL MEETING SUMMARIZER
Shreyas Chaudhari, Aditya Ghamat, Vivek Billa, Chirag Lokhande, Dr. Ramesh Shahabade
DOI: 10.17148/IJARCCE.2026.154241
Abstract: Taking notes at meetings manually can be prone to errors, tedious, and often does not clearly delineate actionable follow-up items for team members. This is especially true in remote collaboration settings that inherently do not facilitate real-time insights. In this work, we present a scalable, web-based Virtual AI Meeting Summarizer that provides transcription, summarization, action item extraction, and upcoming proposed sentiment analysis (if requested), in real time. Our tool combines streaming transcription using Deepgram's high-speed ASR API, concise summarization using Large Language Models (LLMs) accelerated by Groq, and action item detection using rule-assisted NLP. The system utilizes a unified Node.js microservices architecture, WebSocket streaming, and browser-native audio capture via a Chrome extension. The architecture allows for secure, low-latency pipelines, PDF export, and persistent storage via a Node.js user service and PostgreSQL. A review of the literature identifies gaps in meeting summarization technologies with respect to unifying these capabilities in real time or privacy-preserving deployments. Our summarizer fills one or a combination of these gaps through a combined, shareable, extendable approach to real-time summarization in organizational and educational contexts.
Keywords: Meeting Summarization, Action Item Extraction, Whisper Model, Transformer Models, Natural Language Processing, Real-Time Transcription.
Keywords: Meeting Summarization, Action Item Extraction, Whisper Model, Transformer Models, Natural Language Processing, Real-Time Transcription.
Abstract
CAR RENTAL BOOKING SYSTEM USING MERN STACK
Pooja Karche, Namrata Chaple, Komal Kapare, Pratik Shirsath, Prof. S. S Bere and Prof. P. S. Jagtap
DOI: 10.17148/IJARCCE.2026.154242
Abstract: The Car Rental Booking System is a web-based application developed to simplify and automate the process of renting vehicles. The primary objective of this project is to provide users with a seamless platform to search, select, and book cars online, while enabling administrators to efficiently manage vehicle inventory, bookings, and customer data.
Keywords: Car Rental System, MERN Stack, Online Booking
Keywords: Car Rental System, MERN Stack, Online Booking
Abstract
AI-Augmented Business Intelligence Framework for Predictive and Prescriptive Analytics
Asharaf Pulikkalakath, Prasad J C
DOI: 10.17148/IJARCCE.2026.154243
Abstract: The integration of Artificial Intelligence (AI) into Business Intelligence (BI) systems represents a strategic transformation in how businesses interpret data and execute decisions. While traditional BI reports are often static and limiting their ability to solve complex business problems, this paper proposes an AI-Augmented Business Intelligence framework designed to bridge the "decision gap" through predictive and prescriptive analytics. By leveraging ERP data via OData APIs and integrating Machine Learning, Natural Language Processing (NLP), and Agentic AI, the framework delivers real-time insights that improve forecasting accuracy by 15–20%. We examine the advantages of this synergy, including increased operational efficiency, personalized customer experiences, and the role of key tools like Tableau, Power BI, and IBM Watson in enhancing data visualization. the findings indicate that the future of BI is inextricably coupled with AI advancements, enabling organizations to achieve long-term growth and a competitive edge in an increasingly data-driven world.
Keywords: Business Intelligence (BI), Artificial Intelligence (AI), Predictive Analytics, Data-Driven Decision Making, Real-Time Insights
Keywords: Business Intelligence (BI), Artificial Intelligence (AI), Predictive Analytics, Data-Driven Decision Making, Real-Time Insights
Abstract
EXPLAINABLE AI BASED ARRHYTMIA MONITORING SYSTEM
PARIMALA M, VIGNESHWAR P, SAYOOJ KUMAR VS, SABARI KANNAN R, VISHWA RC
DOI: 10.17148/IJARCCE.2026.154244
Abstract: Cardiovascular diseases are a major global health concern, creating a need for intelligent and realtime cardiac monitoring systems. Conventional electrocardiogram (ECG) analysis often depends on specialists and may not provide immediate support in emergency or remote settings. This paper presents CardioSense AI, an explainable real-time ECG monitoring platform that combines embedded sensing, deep learning, automated reporting, and a hospital-grade dashboard for rapid rhythm assessment. The system uses an AD8232 sensor with an ESP32 microcontroller for live ECG acquisition. A one-dimensional convolutional neural network (1D-CNN) directly classifies ECG signals into Normal Rhythm, Bradycardia, Tachycardia, and PVC-type Arrhythmia. To improve transparency, Grad-CAM based Explainable AI (XAI) highlights waveform regions influencing the model prediction. The platform also provides live ECG streaming, alerts, confidence indicators, and simulation support using the MIT-BIH Arrhythmia Database. In addition, an automated report generation module enables downloadable clinical summaries after each session. Experimental results show that the proposed system achieves effective real-time classification with interpretability and practical usability. CardioSense AI offers a scalable solution for telecardiology, bedside monitoring, and preventive healthcare applications.
Abstract
CampusXChange: A Django-Based Smart Student Marketplace Platform for Campus-Specific Peer-to-Peer Commerce
Siddesh Nilak, Omkar Pandit, Nihalveer Singh, Sumit Patil, Avinash Sonule
DOI: 10.17148/IJARCCE.2026.154245
Abstract: The proliferation of digital technology has transformed how students interact and conduct transactions within academic communities. General-purpose classified platforms such as OLX, Quikr, and Facebook Marketplace, while widely adopted, fail to provide the trust mechanisms, campus-level proximity filtering, and student-centric workflows necessary for safe intra-college commerce. This paper presents CampusXChange, a full-stack web application that addresses these gaps by offering a closed, verified digital marketplace exclusively for college students. The system is developed using Django 4.x as the backend framework with a Model-View-Template (MVT) architectural pattern, HTML5/CSS3/JavaScript and Bootstrap 5 for the responsive frontend, and SQLite (development) or PostgreSQL (production) as the relational data store accessed via Django's Object-Relational Mapping (ORM) layer. The platform provides secure user authentication restricted to registered students, category-based product listing and discovery, keyword search with advanced filtering, in-platform buyer–seller messaging, a cart and checkout module, and a comprehensive administrative dashboard. Functional testing confirms that all primary modules perform as intended across desktop and mobile environments. The expected impact includes reduced student expenditure on academic essentials, minimized material waste through peer-to-peer reuse, and a strengthened sense of community within the campus ecosystem. This paper details the motivation, architecture, algorithmic design, implementation, and test results of CampusXChange, demonstrating its viability as a deployable academic marketplace solution.
Keywords: Django, Student Marketplace, Campus Commerce, Peer-to-Peer Platform, Web Application, MVT Architecture, ORM, Student Authentication
Keywords: Django, Student Marketplace, Campus Commerce, Peer-to-Peer Platform, Web Application, MVT Architecture, ORM, Student Authentication
Abstract
PCOD/PCOS HEALTH TRACKER: A SMART SYSTEM FOR MONITORING WOMEN’S HEALTH
Bhagyashree Kajale, Rameshwari Kamble, Archita Khedekar, V.A Bhamre
DOI: 10.17148/IJARCCE.2026.154246
Abstract: Polycystic Ovarian Disease (PCOD) and Polycystic Ovary Syndrome (PCOS) are common hormonal disorders affecting a large number of women, especially in their reproductive age. These conditions often lead to irregular menstrual cycles, weight fluctuations, acne, and other health complications. Due to busy lifestyles and lack of awareness, many women fail to monitor their symptoms regularly. This research proposes a PCOD/PCOS Health Tracker system that helps users record, monitor, and analyze their health data in a structured way. The system focuses on tracking menstrual cycles, symptoms, lifestyle habits, and basic health indicators. By providing a centralized platform, the proposed solution aims to improve awareness, encourage healthy habits, and support early detection of health issues. The system is designed to be user-friendly and accessible, making it suitable for everyday use.
Keywords: PCOD, PCOS, Machine Learning, Healthcare AI, Predictive Analytics, mHealth Application, Logistic Regression, Random Forest
Keywords: PCOD, PCOS, Machine Learning, Healthcare AI, Predictive Analytics, mHealth Application, Logistic Regression, Random Forest
Abstract
Digital Complaint Box System
Jayant Vijay Bhagat, Sankalp Santosh Bhosle, Dipak Ramesh Chavan, Prof. P. S. Shekhar
DOI: 10.17148/IJARCCE.2026.154247
Abstract: Most colleges, offices, and institutions still rely on old complaint-handling methods — paper forms, suggestion boxes, and emails — that are slow, hard to track, and easy to ignore. This paper introduces the Digital Complaint Box System (DCBS), a simple web-based platform that solves these problems. Any user can open the website, fill in a short form, and submit a complaint in seconds. If they do not want to share their identity, they can do so anonymously. Admins get a secure, private dashboard where they can see every complaint, update its progress (Pending, In Progress, or Resolved), and manage everything from one place. The moment an admin changes a status, the user sees the update on their screen automatically — no refreshing needed. The system was built using HTML, CSS, and JavaScript for the front end; Node.js to handle the server logic; and Google Firebase to store data and manage logins. We tested it across eight real-world scenarios — submitting complaints, logging in as admin, going anonymous, checking on a mobile phone, and watching real-time updates — and every test passed. Compared to paper systems, emails, or expensive enterprise tools, DCBS is faster, more transparent, and free to deploy. This paper walks through why the system was needed, how it was designed and built, what the test results showed, and what we plan to add in the future.
Keywords: Complaint Management System; Web Application; Firebase Realtime Database; Node.js; Grievance Resolution; Three-Tier Architecture; Institutional Transparency.
Keywords: Complaint Management System; Web Application; Firebase Realtime Database; Node.js; Grievance Resolution; Three-Tier Architecture; Institutional Transparency.
Abstract
CRIME VIDEO ANALYSIS AND SUMMARIZATION DASHBOARD
Jeevalakshmi K, Santhosh J, Sivabalan M, Srikaran R, Sriram R
DOI: 10.17148/IJARCCE.2026.154248
Abstract: This paper presents a Crime Video Analysis and Summarization Dashboard that leverages Deep Learning, Computer Vision, and Natural Language Processing (NLP) to automate the detection, classification, and summarization of criminal activities captured in surveillance footage. The system employs Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks for real-time anomaly detection and activity recognition in video streams. An intelligent summarization engine condenses lengthy footage into concise, timestamped crime event reports, reducing manual review time significantly. The dashboard provides law enforcement agencies with an intuitive interface to monitor, query, and export summarized incident reports. The proposed system improves investigation efficiency, enhances situational awareness, and supports smarter, data-driven public safety management.
Keywords: Crime Detection, Video Summarization, Deep Learning, CNN, LSTM, Surveillance, Anomaly Detection, NLP, Public Safety, Smart Dashboard
Keywords: Crime Detection, Video Summarization, Deep Learning, CNN, LSTM, Surveillance, Anomaly Detection, NLP, Public Safety, Smart Dashboard
Abstract
ComplaintHub: A Full-Stack Smart Complaint Management System with Real-Time Tracking and Intelligent Resolution Framework
Rishikesh Vedpathak, Aiman Sayyed, Rudra Tikekar, Sahil Tribhuvan, Prof. R. C. Suryawanshi
DOI: 10.17148/IJARCCE.2026.154249
Abstract: A Smart Complaint Management System is a digital platform designed to streamline the process of registering, tracking, and resolving complaints efficiently. This research paper presents ComplaintHub, a MERN stack-based system that integrates modern web technologies to enhance transparency, accountability, and user engagement. The system supports role-based access (User, Staff, Admin), real-time updates, analytics, and secure authentication. Furthermore, it proposes advanced features such as AI-based categorization, SLA-based escalation, and real-time communication. The study highlights system architecture, implementation, enhancements, and future scope. This paper demonstrates how digital transformation improves grievance redressal mechanisms in institutions and organizations.
Keywords: Complaint Management System, MERN Stack, Web Application, MongoDB, Express.js, React, Node.js, Real-Time Tracking, User Feedback System, Grievance Redressal, Role-Based Access Control, Data Analytics, Secure Authentication, REST API, Automation, Notification System.
Keywords: Complaint Management System, MERN Stack, Web Application, MongoDB, Express.js, React, Node.js, Real-Time Tracking, User Feedback System, Grievance Redressal, Role-Based Access Control, Data Analytics, Secure Authentication, REST API, Automation, Notification System.
Abstract
Smart Sanitation Complaint System Using QR Code,GPS Tracking, and Real-Time Image Verification
Ayush Sarvagod, Aniket Tavar, Jishan Pathan, Prof. Kanchanmala More
DOI: 10.17148/IJARCCE.2026.154250
Abstract: Urban sanitation management in Indian municipalities continues to suffer from slow, opaque, and paper-based complaint handling mechanisms that undermine citizen trust and delay resolution. This paper presents the Smart Sanitation Complaint System (SSCS), a lightweight web-based platform that enables citizens to report sanitation issues in under two minutes by scanning a location-specific QR code, capturing photographic evidence through the device camera, and auto-fetching GPS coordinates via the W3C Geolocation API. Complaints are timestamped and routed to the relevant ward authority dashboard in real time, eliminating manual data entry and ambiguous location descriptions. A comparative evaluation against traditional manual systems and existing mobile applications demonstrates that SSCS reduces average complaint registration time from 48–72 hours to under two minutes, improves location accuracy from manual descriptions to ±5 m GPS precision, and raises user satisfaction from 42% to 87% in pilot surveys. The system is implemented entirely using open-web standards (HTML5, CSS3, JavaScript) without requiring application installation, making it accessible on any smartphone. SSCS aligns with the Smart City Mission of India and contributes a replicable, low-cost digital infrastructure model for urban sanitation governance.
Keywords: Smart City, Sanitation, QR Code, GPS Tracking, Complaint Management System, Web Application, Geolocation API, Urban Governance.
Keywords: Smart City, Sanitation, QR Code, GPS Tracking, Complaint Management System, Web Application, Geolocation API, Urban Governance.
Abstract
IoT-Based Water Quality Monitoring System
Amogha Nair, Hitesha Patil, Lavanya Sawant, Adiba Sayyed
DOI: 10.17148/IJARCCE.2026.154251
Abstract: Water quality monitoring is essential for ensuring safe drinking water and maintaining environmental sustainability. Increasing industrialization, urbanization, and agricultural activities have significantly contributed to water pollution, making continuous monitoring a necessity. Traditional water quality monitoring systems rely on manual sampling and laboratory analysis, which are time-consuming, expensive, and do not provide real-time results.
This paper presents an Internet of Things (IoT)-based water quality monitoring system designed to provide continuous and real-time analysis of key parameters such as turbidity and TDS. The proposed system integrates sensors with an ESP32 microcontroller, which processes the collected data and transmits it to a cloud platform using wireless communication. The data is then visualized on a user-friendly dashboard, allowing remote monitoring and analysis.
The system also includes an alert mechanism that notifies users when water quality parameters exceed predefined safe limits, enabling timely intervention and preventing potential health risks. The proposed solution is cost-effective, scalable, and suitable for deployment in both urban and rural environments. Overall, the system improves efficiency, reduces manual effort, and enhances the reliability of water quality monitoring.
Keywords: IoT, Water Quality Monitoring, Sensors, ESP32, Cloud Computing
This paper presents an Internet of Things (IoT)-based water quality monitoring system designed to provide continuous and real-time analysis of key parameters such as turbidity and TDS. The proposed system integrates sensors with an ESP32 microcontroller, which processes the collected data and transmits it to a cloud platform using wireless communication. The data is then visualized on a user-friendly dashboard, allowing remote monitoring and analysis.
The system also includes an alert mechanism that notifies users when water quality parameters exceed predefined safe limits, enabling timely intervention and preventing potential health risks. The proposed solution is cost-effective, scalable, and suitable for deployment in both urban and rural environments. Overall, the system improves efficiency, reduces manual effort, and enhances the reliability of water quality monitoring.
Keywords: IoT, Water Quality Monitoring, Sensors, ESP32, Cloud Computing
Abstract
EDGE AI ENABLED SMART MEDICAL DISPENSING AND PATIENT CARE ROBOT FOR HOSPITAL ROOMS
DR. AARTHY R, NAGALAKSHMI B, NEHA S KUMAR, SUKITHA S
DOI: 10.17148/IJARCCE.2026.154252
Abstract: Edge Artificial Intelligence (Edge AI) and the Internet of Things (IoT) are transforming modern healthcare by enabling real-time, intelligent decision-making directly at the device level while ensuring low latency, reliability, and data privacy. However, existing hospital medication dispensing and patient monitoring systems largely rely on manual operations or cloud-based processing, which can lead to human errors, delayed responses, network dependency, privacy concerns, and increased workload for healthcare staff, limiting their effectiveness in critical care environments.To overcome these challenges, this project proposes an Edge AI-based Smart Medical Dispensing and Patient Care Robot that autonomously operates within hospital rooms. The robot navigates using IR sensor-based line following, verifies patient identity through YOLOv11 for face detection and LBPH for face recognition, and maps the authenticated patient with a locally stored prescription database. It further enables offline voice interaction using the Vosk speech recognition model, allowing the robot to communicate with patients in multiple languages such as English, Tamil, and Malayalam, thereby improving accessibility and user experience across diverse patient groups.In addition, the system performs on- device monitoring of vital parameters such as body temperature and pulse rate. All processing is executed at the edge, and any detected abnormalities are immediately communicated to caregivers through automated alerts. The proposed system significantly reduces medication errors, minimizes dependency on cloud infrastructure, enhances patient safety and data privacy, lowers operational costs, and reduces caregiver workload, making it a scalable and efficient solution for intelligent hospital automation.
Abstract
Sign Language Recognition Using Deep Learning-A Review
Karthik Reddy A, Kushal Reddy K, Mallikarjuna S, P Akash Patil, Dr. Muhibur Rahaman T.R
DOI: 10.17148/IJARCCE.2026.154253
Abstract: For millions of people who are deaf or hard of hearing, sign language is more than just a communication tool it is the very foundation of their identity and daily life. Yet, most of the hearing world cannot understand it, which creates a serious and ongoing barrier to education, employment, and basic social participation. This paper looks at how deep learning can help close that gap through automated Sign Language Recognition (SLR). We reviewed systems that range from simple rule-based approaches all the way to the latest transformer models and graph neural networks. To make sense of all these different methods, we grouped them into a four-tier classification based on how advanced and how deployable they are. We then studied ten important research papers published between 2015 and 2024, compared them across factors like accuracy, speed, and real-world usability, and identified six gaps that still need to be addressed — including the near- total absence of Indian Sign Language datasets and the difficulty of running these systems on everyday mobile devices. Based on all of this, we outline a practical recognition framework called the Deep Sign Recognition Framework (DSRF) that aims to work in real-world settings, support Indian Sign Language, and run on standard hardware without needing expensive equipment.
Keywords: Sign Language Recognition, Deep Learning, CNN, LSTM, Transformer, Hand Gesture, Computer Vision, Accessibility, Indian Sign Language, Transfer Learning.
Keywords: Sign Language Recognition, Deep Learning, CNN, LSTM, Transformer, Hand Gesture, Computer Vision, Accessibility, Indian Sign Language, Transfer Learning.
Abstract
Women Safety Alert System: An Android Application for Real-Time Emergency Response
Sakshi Singh, Trupti Sanap, Mahika Singh, Kalpesh Patil
DOI: 10.17148/IJARCCE.2026.154254
Abstract: The Women Safety Alert System is an Android-based mobile application designed to enhance the personal safety of women by providing immediate emergency assistance during dangerous situations. The application utilizes smartphone technologies such as GPS, SMS services, internet connectivity, sensors, and multimedia recording to create a comprehensive real-time emergency alert platform. The system enables women to quickly notify trusted contacts and emergency responders through a single-touch SOS button or hidden triggers such as shake detection and voice commands. Upon activation, the application fetches the user's real-time location and transmits emergency alerts to predefined contacts via SMS and cloud notifications. The proposed solution integrates multiple emergency communication methods, hidden activation features, real-time GPS tracking, and evidence collection into one unified platform, developed using Android Studio with Java/Kotlin, Firebase, Google Maps API, and Android SDK tools. Experimental testing confirms fast alert transmission, reliable GPS tracking, and effective emergency response support.
Keywords: Women Safety, Android Application, GPS Tracking, SOS Alert System, Emergency Response, Mobile Security.
Keywords: Women Safety, Android Application, GPS Tracking, SOS Alert System, Emergency Response, Mobile Security.
Abstract
DeepScan: A Heuristic-Based Framework for Deepfake and AI-Generated Image Detection Without Neural Network Inference
Neelesh N Shrinivasan, M. Sreedharan, Sriramji P, H. Mary Shiny
DOI: 10.17148/IJARCCE.2026.154255
Abstract: As AI-generated imagery becomes increasingly difficult to distinguish from authentic photographs, the need for accessible detection tools has never been greater. This paper presents DeepScan, a lightweight heuristic-driven image analysis framework that identifies AI-generated or face-swapped images without relying on any pre-trained neural network or GPU hardware. DeepScan applies six calibrated visual heuristics — skin pixel ratio, dark region density, centre-to-background sharpness differential, colour palette diversity, face-region noise estimation, and Error Level Analysis — and combines them through a weighted scoring mechanism to produce a composite authenticity score. The system outputs one of three verdicts: Likely Real, Uncertain, or Likely Fake. Testing across a diverse set of AI-generated portraits and real-world photographs shows a mean fake score of 75.9% for synthetic faces and 8.3% for authentic images, demonstrating strong class separation. Built with Python, Pillow, NumPy, and Flask, DeepScan requires no training phase and consumes minimal computational resources. It is deployed as a REST API accessible through any web browser, making it immediately practical for journalists, media platforms, and content moderation teams.
Keywords: Deepfake Detection, AI-Generated Image Analysis, Error Level Analysis, Heuristic Image Forensics, Skin Pixel Ratio, Colour Diversity, Face Noise Estimation, Image Authenticity, Media Forensics, Flask API, Synthetic Media Detection
Keywords: Deepfake Detection, AI-Generated Image Analysis, Error Level Analysis, Heuristic Image Forensics, Skin Pixel Ratio, Colour Diversity, Face Noise Estimation, Image Authenticity, Media Forensics, Flask API, Synthetic Media Detection
Abstract
“AIVA: An AI-Powered Multi-Functional Virtual Assistant Mobile Application Using Flutter”
Komal Gulve, Kirti Nazirkar, Sanika Veer, Mohini Narwade, Mr. D.S. Jaybhay
DOI: 10.17148/IJARCCE/2026.154256
Abstract: In the past several years, the use of virtual assistants has increased tremendously as a means of enhancing the productivity of users and simplifying their daily activities. However, the majority of current virtual assistance systems typically have limited features, forcing users to use separate applications for many different uses. A common platform to combine multiple functionalities is the mobile application AIVA (Artificial Intelligent Virtual Assistant), which is a multipurpose, artificial intelligence-powered, mobile application that has been developed with the Flutter application framework. AIVA combines an artificial intelligence-based chat interface, an optical character recognition (OCR) feature, a study planner, a reminder system, an advanced scientific calculator, and other related functionalities into a single app. AIVA uses artificial intelligence technologies such as Natural Language Processing (NLP) and Transformer-based AI models (Gemini API) to understand user input and generate intelligent responses. The application will also incorporate OCR to recognize printed text in images and will store and manage data in a format to facilitate the reminders and scheduling functions. Ultimately, this application is designed to provide a unified, user-friendly, and efficient means of minimizing or eliminating the need for multiple applications. The AIVA application has been developed using a modular approach to ensure that it is scalable and efficient for future growth. Final Results: The AIVA application has been demonstrated to successfully integrate a variety of features into one application, providing an improved level of usability, productivity, and experience for the end user.
Keywords: Artificial Intelligence (AI), Virtual Assistant, Natural Language Processing (NLP), Optical Character Recognition (OCR), Flutter Framework, Mobile Application, Task Management, Reminder System, Chatbot, Transformer Model.
Abstract
Sentiment Analysis of Comments Received Through E- Consultation Module Software
Karuturi Shravani, Kavana K, N Suchitra, Preethi K, Dr.Muhibur Rahman T.R
DOI: 10.17148/IJARCCE.2026.154257
Abstract: E-consultation platforms collect a large amount of feedback from users, but going through all these comments manually is difficult and time-consuming. This project focuses on building a sentiment analysis system that can automatically understand and classify user opinions as positive, negative, or neutral. It uses natural language processing (NLP) techniques such as text cleaning, tokenization, and removal of unnecessary words to prepare the data. Machine learning methods like Naïve Bayes and Support Vector Machine are applied to identify the sentiment, and advanced models can be added to improve accuracy when needed. The system also presents the results using simple charts and graphs, making it easier to understand overall user opinion and identify common issues. This helps organizations quickly take better decisions and improve their services. The project also highlights that many existing systems do not fully support real-time analysis, multiple languages, or easy interpretation of results, and aims to provide a more practical and efficient solution.
Keywords: Sentiment Analysis; Natural Language Processing (NLP); Machine Learning; Text Classification; Naïve Bayes; Support Vector Machine (SVM); Logistic Regression; Deep Learning; User Feedback Analysis; E-Consultation Systems; Opinion Mining; Data Visualization; Public Opinion Analysis; Predictive Analytics.
Keywords: Sentiment Analysis; Natural Language Processing (NLP); Machine Learning; Text Classification; Naïve Bayes; Support Vector Machine (SVM); Logistic Regression; Deep Learning; User Feedback Analysis; E-Consultation Systems; Opinion Mining; Data Visualization; Public Opinion Analysis; Predictive Analytics.
Abstract
Optimized Ensemble Intrusion Detection: Balancing Data with SMOTE-ENN and Feature Selection via Jaya Algorithm
Subba Reddy K, Nikhitha Dhayepule
DOI: 10.17148/IJARCCE.2026.154258
Abstract: Network Intrusion Detection Systems (NIDS) are highly significant in ensuring that computer networks are not exposed to emerging cyber threats. However, non-balanced datasets can also reduce the accuracy of detection, particularly on minority attack classes, and non-important features can also obstruct this. This paper proposes a superior way of detecting intrusions within groups. It applies SMOTE-ENN, which equalizes the classes and Jaya Optimization, which selects the most suitable features. To test the system, the NSL-KDD and UNSW-NB15 datasets are used. Pretrained data is trained by individual classifiers such as Decision Tree, Random Forest, ExtraTree, J48 and Bagging with Decision Tree. It is further followed by an ensemble Voting Classifier which composes ExtraTree and Boosted Decision Tree. Tests show that the suggested model works better than others; it gets 100% accuracy, precision, recall, and F1-score on both sets of SMOTE-ENN data, and up to 92% accuracy on unbalanced UNSW-NB15 data. The approach is effective in fixing the imbalance between classes, simplifying calculations, and enhancing the identification of minority types of attacks. This renders it a viable and solid solution to network security challenges in the real world.
Keywords: Network Intrusion Detection System (NIDS), SMOTE-ENN, Jaya Optimization, Feature Selection, Ensemble Learning, Voting Classifier.
Keywords: Network Intrusion Detection System (NIDS), SMOTE-ENN, Jaya Optimization, Feature Selection, Ensemble Learning, Voting Classifier.
Abstract
Blockchain-Based Anti-Counterfeit Product Identification System: A Comprehensive Survey
Alam Basha N, Avinash V K, B K Raghavendra, Dandiya Mohammad Kaif, Dr. Muhibur Rahman T R
DOI: 10.17148/IJARCCE.2026.154259
Abstract: Counterfeit products pose a significant and growing threat to global supply chains, resulting in substantial financial losses, diminished brand equity, and risks to consumer health and safety. Conventional authentication mechanisms depend on centralized databases that remain susceptible to unauthorized alteration and offer limited transparency to downstream stakeholders. Blockchain technology has emerged as a compelling countermeasure, owing to its decentralized architecture, cryptographic security, and append-only immutability. This paper presents a complete survey of blockchain-based anti-counterfeit systems, with particular emphasis on QR code integration, smart contract deployment, and end-to-end product traceability. Let me tell you, a structured fourtier taxonomy is proposed to classify existing systems according to their functional sophistication, from basic barcode verification to fully integrated supply chain solutions. The truth is, critical performance dimensions, including security assurance, scalability, implementation cost, and end-user usability, are examined in depth. A comparative analysis of fifteen representative studies highlights their respective strengths and limitations. The survey further identifies persistent research gaps, including the absence of holistic system integration, susceptibility of QR codes to physical cloning, and constraints on large-scale throughput. Prospective research directions are outlined to guide the development of more strong, scalable, and user-accessible anti- counterfeit solutions.
Keywords: Blockchain; Counterfeit Detection; QR Code; Supply Chain; Smart Contracts; Traceability; Product Authentication; Decentralized Systems.
Keywords: Blockchain; Counterfeit Detection; QR Code; Supply Chain; Smart Contracts; Traceability; Product Authentication; Decentralized Systems.
Abstract
ConformityGate: Intelligent Honeypot-Based Intrusion Detection Framework for Secure IoT Networks
Deepak Marlecha, Zarana J Shah, Rishab P K, Netrang V Davey, Aditi Singh, Ms Charulatha R T
DOI: 10.17148/IJARCCE.2026.154260
Abstract: More gadgets connected online - hospitals, factories, cities - are opening new doors for hackers. Old security tools struggle to keep up. A fresh approach called ConformityGate steps in here. Instead of just blocking attacks, it guides them away quietly. Suspicious behavior triggers a hidden shift. Traffic gets rerouted without warning. Fake setups wait there, watching everything the attacker does. These decoys collect real-time details on how intruders operate. Behind the scenes, Python code runs the show. It talks to sensors scattered through the network. Data streams in nonstop: logins, device status, traffic patterns. When something odd crosses set limits, action follows fast. No loud alarms. Just smooth isolation. Alerts spread like ripples across linked devices. Evidence stays intact. Moves stay unseen. One test after another, through five fake IoT setups, showed it spots attacks 96.7% of the time. False alarms pop up just 2.9% of the time. Responses take an average of 1.6 seconds to kick in. That speed and precision beats older methods relying on signatures, odd behavior patterns, or machine learning models. Built tough but simple, ConformityGate fits real-world needs. It scales without breaking down. Works across many kinds of smart devices. Protection stays strong even when threats get clever. This isn’t theory - it runs where it matters.
Internet of Things security using honeypots and anomaly detection with cyber threat intelligence in network environments through transparent redirection and intrusion detection systems implemented in Python
Internet of Things security using honeypots and anomaly detection with cyber threat intelligence in network environments through transparent redirection and intrusion detection systems implemented in Python
Abstract
AI-Based Smart Forest Safety Monitoring System Using Real-Time Risk Analysis
N. Shahinaz, Nanditha N M, Netravathi Hosamani, K.V. Hima Rashmi, Dr. Muhibur Rahman T.R
DOI: 10.17148/IJARCCE.2026.154261
Abstract: Forest tourism has seen a steady rise in recent years, as more people are drawn to outdoor adventures and nature-based experiences. Despite its popularity, ensuring the safety of tourists in remote and unpredictable forest environments continues to be a major concern. This paper presents a smart forest safety monitoring system that utilizes artificial intelligence to improve user safety through continuous tracking and intelligent risk assessment.
The proposed system integrates GPS-based location tracking with real-time risk evaluation and behavior analysis to provide timely alerts and guidance. It features a user registration module that collects essential personal and medical information, allowing the system to deliver more personalized safety support. In addition, the system is capable of detecting unusual situations, such as extended periods of inactivity or entry into high-risk zones, by analyzing sensor data and location patterns.
During emergency situations, the system can automatically trigger alerts and record short video clips, which can assist in verification and rescue efforts. The system is implemented using modern web technologies, supported by a Flask-based backend for efficient data handling and processing. The results demonstrate that the proposed solution can effectively identify potential risks and respond in a timely manner, making it a practical and reliable approach for enhancing tourist safety in forest environments.
Keywords: Artificial Intelligence (AI), Forest Tourist Safety, Real-Time Risk Assessment, GPS-Based Tracking Systems, Emergency Detection and Response
The proposed system integrates GPS-based location tracking with real-time risk evaluation and behavior analysis to provide timely alerts and guidance. It features a user registration module that collects essential personal and medical information, allowing the system to deliver more personalized safety support. In addition, the system is capable of detecting unusual situations, such as extended periods of inactivity or entry into high-risk zones, by analyzing sensor data and location patterns.
During emergency situations, the system can automatically trigger alerts and record short video clips, which can assist in verification and rescue efforts. The system is implemented using modern web technologies, supported by a Flask-based backend for efficient data handling and processing. The results demonstrate that the proposed solution can effectively identify potential risks and respond in a timely manner, making it a practical and reliable approach for enhancing tourist safety in forest environments.
Keywords: Artificial Intelligence (AI), Forest Tourist Safety, Real-Time Risk Assessment, GPS-Based Tracking Systems, Emergency Detection and Response
Abstract
Dormant Asset Rental for Small Businesses: A Web-Based Digital Marketplace for SME Asset Monetisation
Lalit Sabale, Chaitanya Panmand, Somnath Mulik, Rishabh Pawar, Dr. A R Sonule, Dr. Manoj M Deshpande, Dr. A S Deore
DOI: 10.17148/IJARCCE.2026.154262
Abstract: Small and medium-sized enterprises (SMEs) in India bear a disproportionate financial burden from dormant assets — physical and technological resources that are owned but not continuously utilised. These assets, which include vehicles, machinery, professional equipment, electronics, and event infrastructure, depreciate in value while incurring storage, insurance, and maintenance costs without generating any compensatory revenue. This paper presents Rentify, a full-stack, role-based digital rental marketplace architected to address this problem by enabling SMEs to monetise their underutilised assets through a structured, trusted, and feature-complete online platform. The system is implemented using HTML5, CSS3, and vanilla JavaScript for the frontend presentation layer, Python Flask for the backend REST API, and MySQL 8.0 as the relational database. The platform supports three user roles — renters, vendors, and administrators — each with dedicated dashboards and precisely defined access controls enforced through JSON Web Token authentication. Key features include a multi-category product catalogue, a flexible booking engine with hourly, daily, and weekly pricing tiers, coupon-based discount management, a simulated Razorpay-compatible payment workflow, and a comprehensive transaction history module. The database schema comprises thirteen normalised tables designed to the Third Normal Form. Systematic functional testing across all three user roles confirmed the correctness of all fifty-five tested workflows. API response times averaged 42 milliseconds on a local development server, well within the 200-millisecond threshold for a responsive user experience. The platform directly addresses four research gaps identified in the existing literature: SME neglect in current rental platforms, absence of multi-category solutions, poor identification-to-rental integration, and lack of transparent earning models.
Keywords: Dormant Assets, SME Marketplace, Rental Platform, Flask, MySQL, JWT Authentication, Sharing Economy, Asset Monetisation
Keywords: Dormant Assets, SME Marketplace, Rental Platform, Flask, MySQL, JWT Authentication, Sharing Economy, Asset Monetisation
Abstract
IOT BASED HEALTH CARE PREDICTION USING AI
Rajadurai N, Nagajothi P, Rajalakshmi K, Sridevi M, Yobhashini D
DOI: 10.17148/IJARCCE.2026.154263
Abstract: Wearable health monitoring devices play a crucial role in the timely detection of medical emergencies, such as abnormal heart conditions and falls. But current systems primarily depend on cloud computing, which leads to delays, the need for reliable network connectivity, and privacy issues. To address these issues, this paper proposes an edge- intelligent wearable health monitoring system that can perform real-time health and safety monitoring by fusing multiple sensor readings and applying TinyML. Proposed system incorporates physiological and motion sensors such as heart rate, SpO₂ (oxygen saturation), body temperature and a triaxial accelerometer, interfaces to an ESP32 microcontroller. Rather than sending data to the cloud, the system leverages on-device analytics with TinyML models. This allows real-time detection of cardiac abnormalities and falls to occur on the device. The multi-sensor fusion and embedded machine learning enhance detection performance while minimizing energy and bandwidth costs. The edge computing approach allows for rapid response, improved security and privacy, and does not rely on a strong network connection. In case of a fall or other abnormal event, the device sends immediate notifications to family or health-care providers via wireless communication. The developed system offers a small, low-cost and scalable solution to continuous health monitoring, enabling safe and independent living for the elderly and patients who need continuous monitoring.
Keywords: Wearable-type sensors, IoT healthcare, real-time monitoring, fall detection, edge computing, Tiny ML, low-power systems, embedded system.
Keywords: Wearable-type sensors, IoT healthcare, real-time monitoring, fall detection, edge computing, Tiny ML, low-power systems, embedded system.
Abstract
HistoAssist: A Production-Ready Full-Stack AI Framework Bridging Deep Learning Histopathology and Empathetic Patient Communication
Ali Khan Ayyub Khan, Altaf Ahmed Kasu, Khan Umair Abdul Salam, Md Yusuf Ansari, Alfiya Mulla, Zeeshan Khan
DOI: 10.17148/IJARCCE.2026.154264
Abstract: Artificial intelligence has significant potential in digital pathology, but its practical use is often limited due to issues like secure user access, lack of auditability, and difficulty in explaining results to patients. This work introduces HistoAssist, a ready-to-deploy diagnostic system designed to overcome these challenges. It combines a lightweight CNN built with TensorFlow/Keras to classify histopathology images as benign or malignant, along with a FastAPI backend that provides JWT authentication, secure data handling, and automated report generation. A React.js frontend supports smooth clinical interaction, while a rule-based NLP chatbot explains medical outcomes in simple and empathetic language. HistoAssist goes beyond a research prototype by providing a complete and deployable solution for modern telepathology.
Keywords: Deep Learning, Digital Pathology, Patient-Centred Care, Medical Image Analysis, RESTful Architecture, Artificial Intelligence, Histopathology
Keywords: Deep Learning, Digital Pathology, Patient-Centred Care, Medical Image Analysis, RESTful Architecture, Artificial Intelligence, Histopathology
Abstract
CYCLE CARE: A FLUTTER-BASED MENSTRUAL CYCLE TRACKING AND HEALTH MANAGEMENT APPLICATION
P.S Jogdand, S.S Gharat, H.R Deshmukh, S.V Ankam
DOI: 10.17148/IJARCCE.2026.154265
Abstract: Cycle Care is a cross-platform mobile application developed using Flutter that provides women with a comprehensive menstrual cycle tracking and personal health management system. The application integrates cycle prediction using historical data, daily health logging, water intake tracking with scheduled reminders, dietary insight logging, and multilingual support. Built with a local SQLite database for privacy-first data storage and using shared preferences for user settings, Cycle Care aims to give users meaningful insights into their reproductive health without relying on cloud services. This paper describes the motivation, system architecture, module design, technical implementation, and future scope of the Cycle Care application.
Keywords: Menstrual Cycle Tracker, Flutter, Mobile Health App, Cycle Prediction, Water Reminder, SQLite, Health Logging.
Keywords: Menstrual Cycle Tracker, Flutter, Mobile Health App, Cycle Prediction, Water Reminder, SQLite, Health Logging.
Abstract
Solutions of Burgers' Equation for Modeling Pulsatile Blood Flow in Arteries using Homotopy Analysis Method
Dr. Manoj Yadav*, Prof. Diwari Lal
DOI: 10.17148/IJARCCE.2026.154266
Abstract: The Burgers' equation, a fundamental partial differential equation that combines both non-linear advection and diffusion effects, has been widely utilized to model fluid dynamics in various contexts. In this study, we focus on the solutions of the Burgers' equation to model pulsatile blood flow in arteries, accounting for the complex interplay between the non-linear convective transport and viscous diffusion of blood under pulsatile pressure gradients. We derive both analytical and numerical solutions to the modified Burgers' equation that incorporates a sinusoidal source term to represent the oscillatory nature of blood flow driven by the cardiac cycle. The analytical solutions are obtained using perturbation methods, providing insights into the zeroth-order steady-state flow and higher-order corrections due to pulsatility and non-linearity. The results illustrate the formation of wave-like structures in the velocity profiles, highlighting the impact of varying parameters such as viscosity, pressure gradient frequency, and amplitude on the blood flow patterns. The proposed model offers a simplified yet effective approach to understanding arterial blood flow dynamics, with potential applications in predicting hemodynamic conditions in normal and pathological states. This work underscores the versatility of the Burgers' equation in modeling complex biological flows and contributes to the development of more accurate and efficient models for cardiovascular fluid dynamics.
Keywords: Burgers' Equation, Pulsatile Blood Flow, Arterial Flow Modeling, Sinusoidal Pressure Gradient, Wave-like Structures, Homotopy Analysis Method
Keywords: Burgers' Equation, Pulsatile Blood Flow, Arterial Flow Modeling, Sinusoidal Pressure Gradient, Wave-like Structures, Homotopy Analysis Method
Abstract
Study Material Sharing Platform
Atharva A. Bhosale, Dewank B. Badiger, Sameer Bisen, Adityaraj Godbole, Smita Chunamari
DOI: 10.17148/IJARCCE.2026.154267
Abstract: The Study Material Sharing Platform is a web-based application developed to provide a centralized, secure, and user-friendly environment for managing academic resources. The platform addresses common issues faced by students and faculty, such as scattered study materials, outdated files, poor discoverability, and a lack of quality control. It enables students to upload, download, bookmark, rate, and review materials, while faculty members can organize and manage content efficiently, and administrators can supervise users and resources. The proposed system follows a modern client-server architecture with a React.js front end and Supabase as the back-end-as-a-service platform, providing a PostgreSQL database for structured metadata storage, built-in authentication, and secure cloud file storage. Major features include role-based access control, subject-wise categorization, keyword search and filters, material versioning, notifications, and discussion support. Testing demonstrated strong reliability and usability, with a System Usability Scale (SUS) score of 77.2, placing it well above average usability benchmarks. The project demonstrates how a focused academic repository can improve accessibility, collaboration, and productivity in engineering education.
Keywords: Study Material Sharing, Web Application, React.js, Supabase, PostgreSQL, Role-Based Access Control, Collaborative Learning.
Keywords: Study Material Sharing, Web Application, React.js, Supabase, PostgreSQL, Role-Based Access Control, Collaborative Learning.
Abstract
A Digital Solution to Wrong E-Challan
Suyash S. Salunkhe, Vedika S. Sakharkar, Aditya D. Sharma, Adwait R. Velankar, Rakesh C. Suryawanshi
DOI: 10.17148/IJARCCE.2026.154268
Abstract: The rapid expansion of automated traffic enforcement infrastructure in Indian cities has significantly improved law compliance but has simultaneously exposed a critical systemic flaw: citizens receive e-challans with incorrect vehicle details, wrong violation types, or fraudulent claims with no easy mechanism for verification or dispute. When an innocent citizen receives a wrongly issued e-challan, they face significant legal, financial, and administrative hardship. This paper proposes 'A Digital Solution to Wrong E-Challan', a comprehensive verification and dispute resolution framework engineered to prevent false challan consequences at their root. The system integrates an AI-powered evidence capture module using EasyOCR for number plate recognition with multiple preprocessing strategies (CLAHE, bilateral filtering, Otsu thresholding), a vehicle attribute verification engine using YOLOv8 for vehicle type classification and HSV color space analysis with K-means clustering for cross-validation of vehicle color against a MySQL-backed vehicle registration database containing 30+ sample RC records. A public verification portal allows citizens to authenticate challans without login, and an automated SMS notification module using Fast2SMS API provides real-time citizen communication. A structured dispute resolution workflow enables citizens to report incorrect challans with categorized complaints (wrong vehicle, wrong location, wrong violation, duplicate, fake challan) with status tracking from PENDING through REVIEWED to RESOLVED/REJECTED. The Flask-based backend implements 8 route modules including detect(), generate_challan(), verify_challan(), and report_challan() with session-based authentication and UUID-based verification tokens. Experimental results demonstrate a challan verification accuracy of 91% on plate detection across multiple preprocessing strategies, 87% on HSV-based color matching, and 94% on YOLOv8-based vehicle type classification, with an end-to-end processing latency of under 3 seconds per verification. The proposed system is designed as a software-first, web-accessible solution compatible with existing traffic enforcement infrastructure, making it highly deployable across Indian cities without additional hardware investment.
Keywords: E-Challan, Automatic Number Plate Recognition, EasyOCR, YOLOv8, HSV Color Detection, Fast2SMS, Dispute Resolution, Public Verification, Flask, MySQL, Intelligent Transportation Systems.
Keywords: E-Challan, Automatic Number Plate Recognition, EasyOCR, YOLOv8, HSV Color Detection, Fast2SMS, Dispute Resolution, Public Verification, Flask, MySQL, Intelligent Transportation Systems.
Abstract
AEGIS-X: AI-POWERED AUTONOMOUS SURVEILLANCE AND COMMAND SYSTEM
SHAKIRA BANU L, ME., PARVAZE AHAMED, PRABHU C, SRIMURUGAN B, SHAGUL HAMMED J M
DOI: 10.17148/IJARCCE.2026.154269
Abstract: The rapid growth of digital payment systems and e-commerce platforms has significantly improved convenience but has also led to an increase in online payment frauds and account takeover attacks. Traditional One-Time Password (OTP)-based authentication systems are vulnerable to social engineering attacks such as phishing, fake customer support calls, and deceptive messages. This paper proposes a contextaware security framework that enhances OTP-based authentication by integrating transaction details, location analysis, and real-time cybercrime reporting. The system links OTPs with billing information such as merchant name, transaction amount, and masked card details, enabling users to verify transactions before authorization. Additionally, the system detects suspicious activities using location-based anomaly detection and provides instant fraud reporting with transaction blocking. This approach transforms security from a reactive to a proactive model, reducing financial losses and improving user trust in digital payment systems.
Keywords: OTP Security, Payment Fraud Detection, Context-Aware Authentication, Cybercrime Reporting, Ecommerce Security
Keywords: OTP Security, Payment Fraud Detection, Context-Aware Authentication, Cybercrime Reporting, Ecommerce Security
Abstract
Job Description and Resume Matcher
Harsh Pandurang Korde, Soham Hanumant Shinde, Sumit Rajdev Mourya, Kanchan More
DOI: 10.17148/IJARCCE.2026.154270
Abstract: The Job Description and Resume Matcher is a web-based application developed to simplify and digitize the process of recruitment and candidate selection. Traditional hiring systems rely heavily on manual resume screening and comparison, which are time-consuming and prone to human errors. This project introduces an automated solution that enables recruiters to match candidate resumes with job descriptions efficiently with minimal effort.
The system provides functionalities such as job posting management, resume upload and parsing, skill extraction, automatic matching of qualifications, experience, and keywords, and generation of candidate ranking reports. It ensures data accuracy and consistency while improving productivity. The application is built using the MERN stack, which includes Express.js, React.js, and Node.js, ensuring scalability and performance.
The proposed system is designed to be user-friendly and secure, making it suitable for companies, HR departments, and placement cells. The system significantly reduces manual work, enhances hiring efficiency, and provides a reliable platform for managing recruitment processes digitally.
Keywords: Resume Matching, Web Application, Frontend Development (HTML, CSS, JS), Recruitment System, Candidate Screening, Skill Gap Analysis, Client-Side PDF Parsing, Job Description Analysis
The system provides functionalities such as job posting management, resume upload and parsing, skill extraction, automatic matching of qualifications, experience, and keywords, and generation of candidate ranking reports. It ensures data accuracy and consistency while improving productivity. The application is built using the MERN stack, which includes Express.js, React.js, and Node.js, ensuring scalability and performance.
The proposed system is designed to be user-friendly and secure, making it suitable for companies, HR departments, and placement cells. The system significantly reduces manual work, enhances hiring efficiency, and provides a reliable platform for managing recruitment processes digitally.
Keywords: Resume Matching, Web Application, Frontend Development (HTML, CSS, JS), Recruitment System, Candidate Screening, Skill Gap Analysis, Client-Side PDF Parsing, Job Description Analysis
Abstract
Machine Learning in Career Assistance and Job Application Automation: A Comprehensive Review
Seema, Sanjana R Bharade, Eshwari, Ananya
DOI: 10.17148/IJARCCE.2026.154271
Abstract: Job hunting today is far more complex than it was a decade ago. Candidates must adapt resumes for each role, track multiple applications, and continuously upskill. This paper presents CareerSync, an intelligent system that automates resume parsing, job matching, skill gap analysis, career path recommendation, and job application submission. The system uses BERT-based NLP for accurate resume understanding, along with a hybrid ensemble of Random Forest, SVM, and LSTM models to improve job-candidate matching performance. Additionally, it integrates a recommendation engine to suggest personalized career paths and learning opportunities based on identified skill gaps. A Robotic Process Automation (RPA) module enables seamless interaction with job portals, reducing manual effort in the application process.
Keywords: Career assistance, job recommendation, NLP, BERT, LSTM, resume parsing, RPA
Keywords: Career assistance, job recommendation, NLP, BERT, LSTM, resume parsing, RPA
Abstract
Multi-Agent Orchestration for Enhanced Text-to-SQL Generation: A Schema-Aware Approach with Self-Correction
Vijay M. Rakhade, Parth Gosavi, Chirag Kotkar, Pranav Kashmire, Siddharth Tripathi
DOI: 10.17148/IJARCCE.2026.154272
Abstract: The unification of the functionality of databases and natural language processing Databases has traditionally only provided a means to be able to write SQL queries, however there is an ongoing challenge of converting NL (Natural Language) queries into Structured Query Language(SQL). However, LLMs have significant challenges in handling complex schema structures and long-range dependencies, as well as a tendency toward structural hallucination. This paper presents a novel multi-agent architecture based on LangGraph with the aim to close the gap between NL and SQL through specialized decomposed reasoning. Our implemented system incorporates a robust AST-based Self-Correction loop, a query planning agent for chain-of-thought reasoning, and a semantic-aware Schema Linking Agent using Sentence Transformers. Experimental results on 100 examples from the Spider benchmark show that the complete four-phase system (Model C) achieves the highest Execution Accuracy at 76.0%, outperforming the monolithic zero-shot baseline (63.0%) by +13.0 percentage points. The Schema Linking Agent alone (Model A) achieves 73.0% EX, while the addition of the AST-based Self-Correction loop (Model B) reaches 75.0% EX with an average correction rate of 0.16. Semantic value mapping, dynamic schema discovery, and query complexity routing are identified as crucial paths for future enterprise deployment by a thorough gap analysis.
Keywords: Text-to-SQL, Multi-Agent Systems, Large Language Models, Schema Evolution, Self-Correction, Vector Search, Cross-Database Reasoning.
Keywords: Text-to-SQL, Multi-Agent Systems, Large Language Models, Schema Evolution, Self-Correction, Vector Search, Cross-Database Reasoning.
Abstract
A Review of Optical Character Recognition for Handwritten Hindi Text
Divya D, Bhuvaneshwari V, Shiva Kumar Swamy J, Siddu, Muhibur Rahman T.R, Anita Patil, Dadapeer
DOI: 10.17148/IJARCCE.2026.154273
Abstract: Handwritten text recognition has long been a challenging area within the field of pattern recognition, especially for complex scripts such as Hindi written in the Devanagari script. Unlike printed text, handwritten Hindi exhibits significant variability in writing styles, stroke order, character shapes, and spacing, making Optical Character Recognition (OCR) a difficult problem. Over the past decade, advancements in machine learning, deep learning, and image processing techniques have significantly improved the performance of OCR systems for handwritten scripts.
This paper presents a detailed review of existing approaches for handwritten Hindi text recognition. It explores traditional methods based on feature extraction and classification, as well as modern deep learning approaches such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and hybrid architectures. The study also examines preprocessing techniques, segmentation challenges, and benchmark datasets used in the domain. A structured classification of OCR systems is proposed based on methodology and level of automation.
Further, a comparative analysis is provided considering recognition accuracy, robustness, dataset dependency, and computational complexity. The review highlights that while deep learning models have achieved promising results, challenges such as lack of large annotated datasets, variability in handwriting, and segmentation errors still persist. The paper concludes by identifying research gaps and suggesting future directions for building more accurate, scalable, and real-time OCR systems for handwritten Hindi text.
Keywords: Optical Character Recognition, Handwritten Hindi, Devanagari Script, Deep Learning, CNN, RNN, Image Processing, Pattern Recognition, NLP
This paper presents a detailed review of existing approaches for handwritten Hindi text recognition. It explores traditional methods based on feature extraction and classification, as well as modern deep learning approaches such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and hybrid architectures. The study also examines preprocessing techniques, segmentation challenges, and benchmark datasets used in the domain. A structured classification of OCR systems is proposed based on methodology and level of automation.
Further, a comparative analysis is provided considering recognition accuracy, robustness, dataset dependency, and computational complexity. The review highlights that while deep learning models have achieved promising results, challenges such as lack of large annotated datasets, variability in handwriting, and segmentation errors still persist. The paper concludes by identifying research gaps and suggesting future directions for building more accurate, scalable, and real-time OCR systems for handwritten Hindi text.
Keywords: Optical Character Recognition, Handwritten Hindi, Devanagari Script, Deep Learning, CNN, RNN, Image Processing, Pattern Recognition, NLP
Abstract
Phishing Website Detection using URL-Based Machine Learning for Real-Time Browser Security
Mohammed Zunaid, Puneeth MP, P Abhishek, Rishi Kumar P, Dr. Muhibur Rahaman T.R
DOI: 10.17148/IJARCCE.2026.154274
Abstract: Phishing attacks have emerged as a major cybersecurity concern, where malicious websites imitate legitimate platforms to deceive users into disclosing sensitive information such as passwords, personal data, and banking credentials. Conventional detection techniques, particularly blacklist-based methods, are often ineffective against newly generated and rapidly evolving phishing URLs. To address this limitation, this paper presents a machine learning-based approach for phishing website detection using URL-based feature analysis. The proposed system focuses on extracting key lexical and structural attributes from URLs, including length, presence of abnormal characters, domain-related properties, and suspicious patterns. These features are used to train classification models such as Logistic Regression, Decision Tree, and Random Forest to distinguish between legitimate and phishing websites. The system is designed with the capability to support real-time deployment, making it suitable for integration with browser-based security mechanisms. Experimental evaluation demonstrates improved detection performance in terms of accuracy, precision, and recall. The proposed approach provides an efficient and scalable solution for enhancing user security and mitigating phishing threats in modern web environments.
Abstract
A Secure and Intelligent Drone-Based Healthcare Logistics System with Real-Time IoT Monitoring
Dr. V. Ponniyinselvan, R. Srihariharan, S. SuryaPrakash, S. Poovarasan, K. Sandhuru
DOI: 10.17148/IJARCCE.2026.154275
Abstract: The rapid advancement of drone technology has enabled transformative applications in healthcare, particularly for delivering medical supplies to remote and emergency-affected regions. Conventional transportation systems often suffer from delays due to traffic congestion, inadequate infrastructure, and geographical constraints, limiting timely access to critical resources. This paper proposes a Smart Drone-Based Medical Supply Delivery System integrated with Internet of Things (IoT) technology to enable efficient, secure, and real-time logistics. The proposed system employs GPS-based navigation for autonomous routing, environmental sensors for payload condition monitoring, and wireless communication modules for continuous data transmission. A secure authentication mechanism is incorporated to ensure that only authorized personnel can access delivered medical packages. The system architecture is designed to enhance delivery reliability, minimize response time, and ensure payload safety under varying environmental conditions. Experimental evaluation demonstrates improved delivery efficiency and reduced latency compared to conventional methods, highlighting its potential to enhance healthcare accessibility in critical and time-sensitive scenarios.
Keywords: Unmanned Aerial Vehicle (UAV), Internet of Things (IoT), Medical Supply Delivery, Real-Time Monitoring, Secure Authentication, GPS Navigation, Smart Healthcare System.
Keywords: Unmanned Aerial Vehicle (UAV), Internet of Things (IoT), Medical Supply Delivery, Real-Time Monitoring, Secure Authentication, GPS Navigation, Smart Healthcare System.
Abstract
Agro Vision AI: A Web-Based Intelligent Agricultural E-Commerce Platform
Gopal Patel, Dhannajay Gupta, Kamal Kumar, Dasharath Gupta, Kumar Bibhuti Bhushan
DOI: 10.17148/IJARCCE.2026.154276
Abstract: Agriculture is a key sector for economic growth, but farmers face problems such as low profits, limited market access, and inefficient supply chains. Modern technologies like cloud computing, artificial intelligence, and e-commerce create opportunities to improve traditional farming systems.
This paper presents Agro Vision AI, a web-based platform that directly connects farmers and merchants. The system supports product listing, online payments, order tracking, and delivery management. It removes intermediaries and increases transparency in agricultural trade. The platform uses modern web technologies for scalability and efficiency. Results show faster transactions, better pricing for farmers, and improved user experience.
This paper presents Agro Vision AI, a web-based platform that directly connects farmers and merchants. The system supports product listing, online payments, order tracking, and delivery management. It removes intermediaries and increases transparency in agricultural trade. The platform uses modern web technologies for scalability and efficiency. Results show faster transactions, better pricing for farmers, and improved user experience.
Abstract
Deep Learning Based Facial Emotion Recognition & Intelligent Facial Affect Detection
Anubhav Sharma, Kajal Muania, Muskan Sen, Tanu Pal, Ujjwal Goel, Mr. Rooban Agrawal, Dr. Uruj Jaleel, Dr. Satish Soni
DOI: 10.17148/IJARCCE.2026.154277
Abstract: Facial Emotion Recognition (FER) plays a crucial role in human-computer interaction, enabling machines to interpret human emotions effectively. With the advancement of deep learning techniques, significant improvements have been achieved in recognizing facial expressions with higher accuracy. This paper presents a deep learning-based approach for facial emotion recognition and intelligent facial affect detection using Convolutional Neural Networks (CNNs). The proposed system analyzes facial features from images and classifies emotions such as happiness, sadness, anger, surprise, fear, and neutral. The study evaluates performance using benchmark datasets and demonstrates improved accuracy compared to traditional machine learning methods. Applications include healthcare, security, education, and human-computer interaction systems.
Keywords: Facial Emotion Recognition, Deep Learning, CNN, Computer Vision, Affect Detection, Artificial Intelligence.
Keywords: Facial Emotion Recognition, Deep Learning, CNN, Computer Vision, Affect Detection, Artificial Intelligence.
Abstract
AI-Based Maternal Healthcare Monitoring: A Comprehensive Survey
Adithi R, Akanksh R, Harshitha T, Shaik Khaja, Muhibur Rahman T.R, Anita Patil
DOI: 10.17148/IJARCCE.2026.154278
Abstract: Maternal healthcare monitoring plays a crucial role in reducing maternal mortality and ensuring the well-being of both mother and child. Despite advancements in medical science, many regions, particularly rural and underdeveloped areas, still face challenges such as limited access to healthcare facilities, lack of continuous monitoring, and delayed identification of high-risk pregnancies. In recent years, Artificial Intelligence (AI), Machine Learning (ML), Internet of Things (IoT), and digital healthcare technologies have emerged as powerful tools to address these challenges.
This paper presents a comprehensive survey of existing maternal healthcare monitoring systems that utilize these advanced technologies. The study reviews a wide range of approaches, including machine learning-based risk prediction models, IoT-enabled real-time monitoring systems, mobile health (mHealth) applications, and conversational platforms such as chatbots and IVR systems. A structured classification of these systems is proposed based on their functionality and level of integration.
The paper further provides a comparative analysis of existing solutions in terms of prediction accuracy, accessibility, scalability, and integration capabilities. The survey reveals that while individual technologies have significantly improved maternal healthcare, most existing systems are fragmented and focus on isolated functionalities. A key observation is the absence of a unified system that integrates real-time monitoring, intelligent risk assessment, and accessible user interaction within a single platform.
Finally, the study identifies critical research gaps and highlights future directions for developing scalable, integrated, and inclusive maternal healthcare systems that can effectively address real-world challenges.
Keywords: Maternal Healthcare; Artificial Intelligence; Machine Learning; IoT; Risk Prediction; Healthcare Monitoring; Natural Language Processing; Digital Health Systems; mHealth; Telemedicine.
This paper presents a comprehensive survey of existing maternal healthcare monitoring systems that utilize these advanced technologies. The study reviews a wide range of approaches, including machine learning-based risk prediction models, IoT-enabled real-time monitoring systems, mobile health (mHealth) applications, and conversational platforms such as chatbots and IVR systems. A structured classification of these systems is proposed based on their functionality and level of integration.
The paper further provides a comparative analysis of existing solutions in terms of prediction accuracy, accessibility, scalability, and integration capabilities. The survey reveals that while individual technologies have significantly improved maternal healthcare, most existing systems are fragmented and focus on isolated functionalities. A key observation is the absence of a unified system that integrates real-time monitoring, intelligent risk assessment, and accessible user interaction within a single platform.
Finally, the study identifies critical research gaps and highlights future directions for developing scalable, integrated, and inclusive maternal healthcare systems that can effectively address real-world challenges.
Keywords: Maternal Healthcare; Artificial Intelligence; Machine Learning; IoT; Risk Prediction; Healthcare Monitoring; Natural Language Processing; Digital Health Systems; mHealth; Telemedicine.
Abstract
Postpartum Depression Risk Prediction Model: Leveraging Machine Learning for Early Detection and Preventive Care
Tanish Shingade, Abhinav Ragam, Shriddhar Patil, Nikhil Nikam, Prof. Smita Chunamari
DOI: 10.17148/IJARCCE.2026.154279
Abstract: Postpartum depression (PPD) is one of the most common and underdiagnosed psychiatric disorder to affect women after giving birth. It not only impacts the mental well-being of the mother but also infant health and development. Despite its seriousness, early detection and treatment remain challenging in the majority of healthcare systems due to the subjective symptoms and the stigma of mental health.
In this project, we applied machine learning to develop a predictive model that can assess whether a woman is at risk for PPD based on clinical, psychological, and lifestyle variables. The dataset contains 1,500 records, of which 1,200 are used for training purposes and 300 for testing purposes. It has variables such as mother's age, number of children, marriage status, history of mental illness, hormone level, stress level, sleeping habits, mode of delivery, and week postpartum.
Two models were used and compared: Logistic Regression and Random Forest Classifier. The Random Forest model achieved a remarkable test accuracy of 100%, but Logistic Regression achieved an accuracy of approximately 81%, a more interpretable and generalizable baseline. The models were measured using standard metrics such as accuracy, precision, recall, F1-score, and confusion matrix.
In order to cross-verify against overfitting, the Random Forest model's excellent accuracy, a 5-fold cross-validation has been performed. The Random Forest model itself had an extremely high mean accuracy of 99.83%, indicating a good generalization ability. Importance analysis of features also indicated that support systems, stress levels, and hormone levels were among the most significant predictor variables.
This work also shows promise for machine learning as a complementary diagnostic tool for postpartum menal health screening, providing clinicians and public health practitioners data-informed guidance to actively identify and support at-risk individuals.
Keywords: Maternal mental health, New born & Mother care, Postnatal health, Lactational Amenorrhea Method (LAM), Postpartum recovery support.
In this project, we applied machine learning to develop a predictive model that can assess whether a woman is at risk for PPD based on clinical, psychological, and lifestyle variables. The dataset contains 1,500 records, of which 1,200 are used for training purposes and 300 for testing purposes. It has variables such as mother's age, number of children, marriage status, history of mental illness, hormone level, stress level, sleeping habits, mode of delivery, and week postpartum.
Two models were used and compared: Logistic Regression and Random Forest Classifier. The Random Forest model achieved a remarkable test accuracy of 100%, but Logistic Regression achieved an accuracy of approximately 81%, a more interpretable and generalizable baseline. The models were measured using standard metrics such as accuracy, precision, recall, F1-score, and confusion matrix.
In order to cross-verify against overfitting, the Random Forest model's excellent accuracy, a 5-fold cross-validation has been performed. The Random Forest model itself had an extremely high mean accuracy of 99.83%, indicating a good generalization ability. Importance analysis of features also indicated that support systems, stress levels, and hormone levels were among the most significant predictor variables.
This work also shows promise for machine learning as a complementary diagnostic tool for postpartum menal health screening, providing clinicians and public health practitioners data-informed guidance to actively identify and support at-risk individuals.
Keywords: Maternal mental health, New born & Mother care, Postnatal health, Lactational Amenorrhea Method (LAM), Postpartum recovery support.
Abstract
A MERN-Stack Blog Application with Gemini-3-Flash Integration
Aman Narayan, Abhinav Nigam, Abhishek Jaiswal, Suraj Kushwaha
DOI: 10.17148/IJARCCE.2026.154280
Abstract: The exponential growth of digital content has necessitated the evolution of traditional Content Management Systems (CMS). Traditional platforms lack integrated intelligent tools, forcing authors to spend significant time on content formulation and moderation. This project presents the development of an intelligent, automated, and highly interactive Blog Application built on the MERN stack (MongoDB, Express.js, React.js, Node.js). The system's core innovation is the integration of the Google Gemini-3-Flash-Preview Generative AI model, which automatically synthesizes high-quality, formatted blog descriptions based on minimal input (Title and Subtitle).Furthermore, the application incorporates a "1-to-N" user architecture featuring a centralized Admin Dashboard. The dashboard provides real-time statistics, comprehensive blog management, and a strict comment moderation workflow to prevent spam. On the client side, the application offers dynamic category filtering, seamless navigation, and social media integration without page reloads. The study demonstrates that integrating Gemini into web frameworks significantly reduces administrative workload while enhancing content quality and platform security.
Keywords: Generative AI, Gemini-3-Flash, MERN Stack, Content Management System (CMS), Comment Moderation, Single Page Application (SPA), Web Security.
Keywords: Generative AI, Gemini-3-Flash, MERN Stack, Content Management System (CMS), Comment Moderation, Single Page Application (SPA), Web Security.
Abstract
“A Study on the Impact of Digital Banking Adoption on Customer Satisfaction and Financial Performance in Pune”
Atharva Somnath Kadam, Laxman Kalke, Dr. Pradyuman Shastri
DOI: 10.17148/IJARCCE.2026.154281
Abstract: These Research paper study the rapid advancement of digital technology has significantly transformed the banking sector, particularly in urban areas like Pune. This study examines the impact of digital banking adoption on customer satisfaction and the financial performance of banks. Digital banking services, including mobile banking, internet banking, and online transactions, offer convenience, speed, and accessibility to customers. However, challenges such as security concerns, technical issues, and lack of digital literacy continue to influence user experience. The research adopts a descriptive and analytical approach to evaluate customer perceptions and the extent of digital adoption. Primary and secondary data are used to analyze the relationship between digital banking usage, customer satisfaction, and bank performance. The findings suggest that increased adoption of digital banking positively affects customer satisfaction and enhances operational efficiency and financial outcomes for banks, although addressing security and awareness issues remains crucial.
Keywords: Digital Banking, Customer Satisfaction, Financial Performance, Technology Adoption, Online Banking, Mobile Banking, Banking Efficiency, Pune, Customer Experience, Banking Innovation
Keywords: Digital Banking, Customer Satisfaction, Financial Performance, Technology Adoption, Online Banking, Mobile Banking, Banking Efficiency, Pune, Customer Experience, Banking Innovation
Abstract
Spatially-Gated CNN-Transformer Hybrids for Pneumonia Classification: A Unified Framework for Metric-Optimized Local-Global Explainability
Ch Mydhili*, B Madhav Rao
DOI: 10.17148/IJARCCE.2026.154282
Abstract: The deep learning paradigm shift is completely transforming image classification especially in the medical scenario where proper classification helps establish the patients' treatment protocol. CNN has proven to be very successful in learning local spatial features such as textures and edges, whereas Swin-T excel in capturing global context dependencies through self-attention mechanism. However, it is evident that both methods have some shortcomings in case of considering only one approach since CNN fails to adequately capture the long-range dependency while Transformer models require huge amounts of data and powerful computations. Although hybrid CNN-Transformer architectures provide an answer to each architecture's limitation by combining their feature learning abilities into one, they remain black-box models which produce hard-to-explain predictions and hence cannot be trusted especially in the medical field where accurate classifications are required. This project aims at finding an answer to this problem by introducing an innovative model called Spatially-Gated CNN-Transformer Hybrid with Dual-Level Explainability. The model will consist of a ResNet-50 CNN backbone to learn local features, and a Swin Transformer to model global context through spatial attention gating mechanism. Besides, this research will introduce a dual-level explainability method involving both Grad-CAM and Attention Rollout.
Keywords: CNN–Transformer Hybrid Network, Dual-Level Explainability, Explanation Fusion, Quality Evaluation Metrics.
Keywords: CNN–Transformer Hybrid Network, Dual-Level Explainability, Explanation Fusion, Quality Evaluation Metrics.
Abstract
Automated Misinformation Detection in Online News Using NLP and LSTM-Based Deep Learning
M Sasidhar*, M Krishna
DOI: 10.17148/IJARCCE.2026.154283
Abstract: The exponential growth of internet-based communication channels has revolutionized information accessibility, yet it has simultaneously intensified the circulation of deceptive, manipulated, and non-credible news content across digital platforms. The uncontrolled diffusion of misinformation can influence public perception, distort democratic processes, trigger financial instability, and create widespread social confusion, making automated verification systems increasingly essential in the modern data ecosystem. This project, titled Automated Misinformation Detection in Online News Using NLP and LSTM-Based Deep Learning, proposes an intelligent text analytics framework that leverages Natural Language Processing and deep sequential learning to distinguish authentic news articles from fabricated narratives. The system begins with structured dataset acquisition followed by extensive preprocessing operations such as text normalization, noise elimination, token segmentation, stop-word filtering, vocabulary encoding, and sequence standardization to transform raw textual inputs into computationally meaningful representations. An embedding layer is employed to learn semantic associations between words, after which a Bidirectional Long Short-Term Memory network captures contextual dependencies from both preceding and succeeding directions within a sentence, enabling superior understanding of narrative flow, linguistic irregularities, and deceptive writing patterns. The model is trained using labeled news corpora and evaluated through rigorous performance indicators including accuracy, precision, recall, F1-score, and confusion matrix interpretation, where the experimental results demonstrate strong predictive capability and robust generalization on previously unseen samples. Compared with conventional machine learning classifiers, the proposed architecture offers enhanced contextual comprehension and higher classification reliability for complex textual data. The modular design of the framework also supports practical deployment in news authentication portals, content moderation systems, browser-based verification tools, and large-scale media monitoring environments. Furthermore, the solution can be extended through multilingual processing, transformer integration, explainable artificial intelligence mechanisms, and real-time misinformation surveillance. Overall, the project establishes that the fusion of advanced NLP techniques with LSTM-driven deep learning forms a scalable, accurate, and future-ready approach for strengthening trust and credibility within digital information networks.
Keywords: Deep Learning, Detection, Accuracy and F-1 Score.
Keywords: Deep Learning, Detection, Accuracy and F-1 Score.
Abstract
MACHINE LEARNING BASED TRAVEL MANAGEMENT AND OPTIMIZATION SYSTEM
Aditya Kumar Yadav, Rohit Yadav, Nishant Singh, Shivam Verma, Ms. Deepika Pandey
DOI: 10.17148/IJARCCE.2026.154284
Abstract: The rapid growth of the tourism industry has increased the demand for intelligent and integrated travel planning systems that can efficiently manage multiple travel services within a single platform. This paper presents Pack For Ride, a machine learning-based travel management and optimization system designed to provide a unified solution for planning complete tours, including transportation, accommodation, food, and other essential services.
The proposed system incorporates map-based distance calculation techniques to accurately estimate travel distance and dynamically predict travel costs. A supervised machine learning model is utilized for cost prediction, where input parameters such as distance, number of travelers, and selected facilities are used to generate optimized and user- specific pricing. This approach improves accuracy and flexibility compared to traditional static pricing methods.
In addition, the system integrates real-time weather information using public APIs to assist users in making informed travel decisions. A recommendation mechanism is also implemented to suggest suitable and optimized travel packages based on user preferences and constraints. The platform is developed using modern web technologies, ensuring scalability, responsiveness, and efficient data processing. It also includes notification services and secure online payment integration to provide a complete end-to-end travel management experience.
Experimental evaluation indicates that the proposed system improves planning efficiency, reduces manual effort, and enhances user experience by delivering a personalized, cost-effective, and intelligent travel solution.
Keywords: Travel Management System, Machine Learning, Cost Prediction, Map-Based Distance Calculation, Weather API, Smart Tourism.
The proposed system incorporates map-based distance calculation techniques to accurately estimate travel distance and dynamically predict travel costs. A supervised machine learning model is utilized for cost prediction, where input parameters such as distance, number of travelers, and selected facilities are used to generate optimized and user- specific pricing. This approach improves accuracy and flexibility compared to traditional static pricing methods.
In addition, the system integrates real-time weather information using public APIs to assist users in making informed travel decisions. A recommendation mechanism is also implemented to suggest suitable and optimized travel packages based on user preferences and constraints. The platform is developed using modern web technologies, ensuring scalability, responsiveness, and efficient data processing. It also includes notification services and secure online payment integration to provide a complete end-to-end travel management experience.
Experimental evaluation indicates that the proposed system improves planning efficiency, reduces manual effort, and enhances user experience by delivering a personalized, cost-effective, and intelligent travel solution.
Keywords: Travel Management System, Machine Learning, Cost Prediction, Map-Based Distance Calculation, Weather API, Smart Tourism.
Abstract
DEVELOPMENT OF A CONVOLUTIONAL NEURAL NETWORK FOR BINARY IMAGE CLASSIFICATION OF BEARS AND PANDAS WITH APPLICATIONS IN DIGITAL IMAGE PROCESSING
P Aarti Pai, Khushi, Sudarshan SR, H Mary Shyni
DOI: 10.17148/IJARCCE.2026.154285
Abstract: The study investigates the design, application, and evaluation of a CNN model that is trained to perform binary classification based on bear and panda images. The use of an open-source dataset allows integrating certain basic ideas behind digital image processing together with deep learning techniques, which can be applied during image preprocessing and classification. The proposed CNN model comprises convolution and pooling layers followed by fully connected layers that can be used for binary classification. The Adam optimizer along with the binary cross entropy loss function were used, whereas accuracy for training the model, precision, and recall were used to measure the performance of the developed model.
Keywords: Image Processing, Binary Image Classification, Convulutional Neural Network
Keywords: Image Processing, Binary Image Classification, Convulutional Neural Network
Abstract
Revolutionizing Precision Agriculture with Machine Learning: Current Progress and Future Directions
Tajbir Singh, Satinder Kaur, Satveer Kour, Sandeep Kaur
DOI: 10.17148/IJARCCE.2026.154286
Abstract: Precision agriculture represents a sophisticated farming paradigm designed to boost productivity while optimizing resource utilization, including water, soil nutrients, and energy. The exponential growth of sensing technologies, Internet of Things devices, drones, and satellite monitoring systems has produced massive agricultural datasets, demanding advanced analytical techniques. Within this framework, machine learning has attracted considerable interest due to its capacity to analyze intricate data and support data-informed decisions in contemporary agriculture. This review offers a thorough examination of ML methods and their roles in precision agriculture, scrutinizing prevalent strategies—such as supervised, unsupervised, deep, and ensemble learning—for processing agricultural data. Principal applications addressed encompass crop yield forecasting, disease identification, intelligent irrigation, soil condition assessment, weed and pest management, and crop surveillance via drones and satellites. The analysis also emphasizes cutting-edge developments in deep learning frameworks, remote sensing tools, and smart monitoring systems, illustrating persistent progress in intelligent agriculture. Nevertheless, obstacles remain in domains like data scarcity, heterogeneous data fusion, computational demands, model transparency, and broad-scale implementation. The review additionally investigates prospective avenues, including explainable artificial intelligence, edge computing, reinforcement learning, and self-governing farming systems. In summary, ML-enabled precision agriculture paves a viable route toward resilient, resource-efficient agricultural systems capable of addressing impending food security imperatives.
Keywords: Precision Agriculture, Machine Learning, Crop Yield Prediction, Agricultural Systems
Keywords: Precision Agriculture, Machine Learning, Crop Yield Prediction, Agricultural Systems
Abstract
Artificial Intelligence: Revolutionizing the Future of Medicine: Intelligent Systems, Deep Learning, and the Next Generation of Clinical Care
Nitin kumar, Jatin kumar, Shivam, Kanika Prajapati, Rooban Aggarwal, Satish kumar soni, Uruj jaleel
DOI: 10.17148/IJARCCE.2026.154287
Abstract: The rapid advancement of artificial intelligence (AI) technologies has precipitated a paradigm shift in healthcare delivery, clinical diagnostics, and patient management. This study investigates the intersection of machine learning, deep learning, and biomedical data science, examining how AI methodologies can be leveraged to enhance diagnostic accuracy, accelerate drug discovery, and personalize treatment protocols. Through comprehensive analysis of electronic health records (EHR), medical imaging datasets, and genomic data, this research demonstrates that AI- powered diagnostic systems achieve accuracy rates of 90-97% [1][2] across multiple disease categories, surpassing traditional clinical benchmarks in several domains. We present an integrated AI framework for clinical decision support that incorporates convolutional neural networks for image analysis, natural language processing for EHR mining, and reinforcement learning for treatment optimization. The findings reveal significant opportunities for reducing diagnostic errors, improving patient outcomes, and optimizing healthcare resource allocation. This study contributes to the growing field of medical AI by providing actionable insights for healthcare organizations, clinicians, and policymakers seeking to responsibly integrate AI into clinical practice while maintaining patient safety, data privacy, and ethical standards.
Abstract
Design of an Integrated Model for Neuro Symbolic GenAI and RAG Driven Personalized Cardiometabolic Care Sets
Sagar Fernandes, Dr. Sangeeta Vhatkar
DOI: 10.17148/IJARCCE.2026.154288
Abstract: Increasing demand for clinically trustworthy, patient-specific decision systems has highlighted a gap between GenAI models and cardiometabolic care, where symptoms fluctuate gradually, data streams change hourly, and therapies must be ethical. Fragmented multimodal integration, shallow personalization, and confident but unsupported suggestions plague existing techniques. Due to these limitations, clinicians may generate technically impressive work that is hard to justify or apply to everyday hypertension and type 2 diabetes treatment. GenAI + RAG learns from heterogeneous data and follows causal, ethical, and evidentiary paths. It solves long-standing difficulties using a context-aware multimodal learning engine, retrieval-grounded generative reasoning, and five analytical methodologies in its validation pipeline. Salt sensitivity and nighttime heart-rate variability are examined for persistence using Multiscale Causal Uncertainty Stratification Analysis. Dynamic Ethical Constraint Verification Engine then assesses ethical restrictions' effects. Hierarchical Evidential RAG Stress-Test Simulation evaluates retrieval under conflicting clinical guidelines. A Persona- Calibrated Recommendation Consistency Audit checks for lifestyle or medication inconsistencies among patient archetypes. The last layer, Longitudinal Personalized Improvement Prediction Benchmark, predicts six-month A1C, systolic blood pressure, and marker trajectories from these suggestions. Causality, contextual explainability, and long- term clinical relevance appear to improve with the architecture. It may imply a generation of medical AI systems that reason modestly, trace their conclusions, and offer clinicians evidence-based pathways rather than isolated predictions.
Keywords: GenAI, RAG Systems, Cardiometabolic Care, Multimodal Learning, Explainable AI, Analysis.
Keywords: GenAI, RAG Systems, Cardiometabolic Care, Multimodal Learning, Explainable AI, Analysis.
Abstract
Advancements in Data Analytics: A Framework for Research Applications
P. Hemalatha
DOI: 10.17148/IJARCCE.2026.154289
Abstract: The rapid growth of data generated from diverse sources has significantly increased the importance of data analytics in research. This paper presents a comprehensive overview of the evolution of data analytics and its role in handling large and complex datasets. It begins by examining the challenges associated with big data and highlights the importance of data engineering in managing and processing such data efficiently. The study further explores key stages of data preparation, including data cleaning, transformation, and modeling.Various analytical approaches are discussed, including exploratory data analysis, predictive modeling, and machine learning techniques, which enable researchers to extract meaningful insights. In addition, the paper reviews commonly used programming languages and tools that support data analytics processes. Recent advancements in the field, such as big data technologies, cloud computing, and data privacy considerations, are also addressed.Overall, this paper aims to provide a structured understanding of data analytics for research applications, serving as a useful resource for researchers seeking to leverage data-driven methods to enhance their studies.
Keywords: Data Analytics, Artificial Intelligence, Machine Learning, Big Data, Cloud Computing, Data Privacy, IoT
Keywords: Data Analytics, Artificial Intelligence, Machine Learning, Big Data, Cloud Computing, Data Privacy, IoT
Abstract
INTELLION BOT AI: FUTURE - READY INTELLIGENT BOT SYSTEM
Laxmi yadav, Harshita Singh, Anshika Yadav
DOI: 10.17148/IJARCCE.2026.154290
Abstract: The rapid growth of intelligent conversational system has led to increased demand for AI-Powered chatbots that can simulate human-like interaction. This paper introduces Intellion Bot AI ,a machine learning-enhanced chatbot built on the MERN technology stack. The system integrates natural language processing models to analyse user queries ,extracts intent, and generate meaningful responses dynamically. MongoDB is used for efficient data storage, while Node.js and express.js handle backend processing and API communication. React.js provides an interactive and responsive user interface.
This paper proposed to automate query resolution, improve communication efficiency, and reduce dependency on manual supports system. Experimental results demonstrates that the system achieves improved contextual understanding and faster response generation. This research highlights the effectiveness of combining ai techniques with full-stack web development to create intelligent, scalable conversational agents.
This paper proposed to automate query resolution, improve communication efficiency, and reduce dependency on manual supports system. Experimental results demonstrates that the system achieves improved contextual understanding and faster response generation. This research highlights the effectiveness of combining ai techniques with full-stack web development to create intelligent, scalable conversational agents.
Abstract
CrackIt: An AI-Powered Interview Preparation and Performance Analysis Platform
Kartik Saini, Shivam Joshi, Mayank Sharma, Khushal Singh, Vishal Singh
DOI: 10.17148/IJARCCE.2026.154291
Abstract: In today’s competitive job market, effective interview preparation plays a critical role in improving employment opportunities for candidates. Traditional preparation methods such as static question banks or manual mock interviews often fail to provide personalized feedback and real-time performance evaluation. This research presents CrackIt, an AI-powered interview preparation platform designed to simulate realistic interview environments using advanced web technologies and Large Language Models (LLMs). The system integrates resume analysis, AI-driven interview simulations, performance analytics, and automated report generation into a unified platform. CrackIt leverages modern frameworks including React, FastAPI, Supabase, and Groq AI models to deliver adaptive interview experiences tailored to individual users. The platform also incorporates secure authentication, interactive dashboards, and downloadable performance reports to enhance user engagement and tracking. The proposed solution aims to bridge the gap between theoretical preparation and real-world interview scenarios by providing intelligent feedback and personalized recommendations. Experimental implementation demonstrates that the platform effectively improves interview readiness through continuous evaluation and structured insights.
Keywords: Artificial Intelligence, Interview Simulation, Resume Analysis, Large Language Models, Web Applications, Performance Analytics
Keywords: Artificial Intelligence, Interview Simulation, Resume Analysis, Large Language Models, Web Applications, Performance Analytics
Abstract
Smart Agriculture System Using Internet of Things
Nikita Chauhan, Mohd. Sarim, Krishan kumar, Harshita, Khushi, Usha Kumari, Satish kumar soni, Uruj Jaleel
DOI: 10.17148/IJARCCE.2026.154292
Abstract: The Internet of Things (IoT) has emerged as a transformative technology for modernizing traditional agricultural practices into smart, data-driven systems. This paper presents a comprehensive review of Smart Agriculture Systems (SAS) leveraging IoT technologies to enhance crop productivity, optimize resource utilization, and enable real- time field monitoring. The proposed architecture integrates heterogeneous IoT sensors including soil moisture sensors, temperature and humidity sensors, pH sensors, and light intensity sensors with cloud computing platforms and machine learning algorithms to provide actionable insights for farmers. Key components such as automated irrigation, crop disease detection, livestock monitoring, and weather-based advisory systems are systematically analyzed. The paper evaluates communication protocols including MQTT, LoRaWAN, Zigbee, and Wi-Fi for agricultural deployments, assessing their suitability based on range, power consumption, and data rate. Experimental results from pilot deployments demonstrate up to 35% reduction in water consumption, 28% improvement in crop yield, and significant reduction in manual labor through automation. Challenges including connectivity in rural areas, interoperability, data security, and high deployment costs are discussed along with mitigation strategies. The paper concludes by examining future directions in AI-integrated smart farming, edge computing, drone-assisted monitoring, and the role of blockchain for supply chain transparency in precision agriculture.
Keywords: Internet of Things (IoT), Smart Agriculture, Precision Farming, Automated Irrigation, Soil Monitoring, Crop Disease Detection, MQTT, LoRaWAN, Edge Computing, Precision Agriculture.
Keywords: Internet of Things (IoT), Smart Agriculture, Precision Farming, Automated Irrigation, Soil Monitoring, Crop Disease Detection, MQTT, LoRaWAN, Edge Computing, Precision Agriculture.
Abstract
A SYSTEMATIC REVIEW OF AI POWERED FOOD RECOGNITION SYSTEM
Harsh Pachauri, Megha Rajput, Krishna Kaushik, Dharmesh Rawat, Pankaj Saraswat, Alok Singh Jadaun
DOI: 10.17148/IJARCCE.2026.154293
Abstract: More people face health issues tied to poor eating habits - conditions like obesity and heart problems keep rising. Because of this shift, better tools for tracking what we eat have become essential. Old ways of recording meals, such as handwritten logs or trying to remember everything eaten in a day, tend to be unreliable. Mistakes happen. Memories fade. Sticking with those methods over time? Rare. A new approach enters here - not magic, just smart engineering. This method uses artificial intelligence to turn meal photos into useful nutrition facts without guesswork. At its core sits a neural network model called ResNet-18, good at telling one dish apart from another. It learned from more than 100,000 food images labeled across 101 categories. Training happened using Food-101, a large-scale collection built for exactly this purpose. Speed matters; PyTorch handles number crunching while OpenCV prepares raw pictures for analysis. Cropping, color adjustments, noise cleanup - all done before classification kicks in. Once identified, each food links to stored nutrient values: calories, protein levels, fat amounts, carbs included. Results appear instantly through a live website powered by Streamlit. Snap a photo, get details moments later. No waiting. No spreadsheets. Starting off different, recent studies - especially those on visionlanguage systems and image-focused workout tools - show our method handles speed and accuracy without favoring one too much. Early tests suggest this automatic system cuts down the hassle of logging by hand, providing something sturdy and flexible enough to grow with personal wellness needs while supporting clearer food choices.
Though not perfect, it fits well where quick results meet reliable detection.
Keywords: Food Recognition, Deep Learning, ResNet-18, Nutritional Analysis, Computer Vision, PyTorch, Calorie Estimation, Food-101 Dataset
Though not perfect, it fits well where quick results meet reliable detection.
Keywords: Food Recognition, Deep Learning, ResNet-18, Nutritional Analysis, Computer Vision, PyTorch, Calorie Estimation, Food-101 Dataset
Abstract
Research Design Approaches in Cybersecurity Studies: A Comprehensive Review of Methods, Challenges, and Future Directions
Kamsali Aishwarya, Keerthi h, Keerthi somaraddi, M Charan, Dr. Muhibur Rahman T.R
DOI: 10.17148/IJARCCE.2026.154294
Abstract: Cybersecurity has emerged as one of the most critical domains in the digital era due to the increasing frequency and sophistication of cyber threats affecting governments, organizations, and individuals. Effective cybersecurity research requires well-structured research designs to ensure reliable findings, practical solutions, and scientific validity. This paper presents a comprehensive review of research design approaches used in cybersecurity studies, including quantitative, qualitative, experimental, and mixed-method methodologies. The study examines common application areas such as network security, malware detection, phishing analysis, privacy protection, and human factors in cybersecurity. It also highlights major challenges faced by researchers, including limited access to real-world attack data, ethical constraints, rapidly evolving threats, and reproducibility issues. Furthermore, the paper identifies current research gaps and proposes future directions involving artificial intelligence, IoT security, cloud security, and privacy-preserving research frameworks. The findings indicate that selecting an appropriate research design significantly improves the quality, relevance, and impact of cybersecurity studies. This review aims to support students, researchers, and practitioners in developing robust methodologies for future cybersecurity investigations.
Keywords: Cybersecurity, Research Design, Quantitative Research, Qualitative Research, Mixed Methods, Experimental Studies, Network Security, Malware Analysis, Phishing Detection, Data Privacy, Ethical Hacking, Artificial Intelligence, IoT Security, Cloud Security, Risk Assessment.
Keywords: Cybersecurity, Research Design, Quantitative Research, Qualitative Research, Mixed Methods, Experimental Studies, Network Security, Malware Analysis, Phishing Detection, Data Privacy, Ethical Hacking, Artificial Intelligence, IoT Security, Cloud Security, Risk Assessment.
Abstract
Data Visualization Techniques and Actionable Insights
Ansh Tyagi, Arpit Tyagi, Aditya Adhana, Ms. Tanya Jain, Dr. Uruj Jaleel, Dr. Satish Kumar Soni
DOI: 10.17148/IJARCCE.2026.154295
Abstract: Data visualization stands at the intersection of art and science, transforming raw numerical data into comprehensible graphical representations that support faster, more accurate decision- making. In an era defined by information overload, organizations that harness the power of effective visualization gain a decisive competitive advantage. This paper presents a comprehensive survey of data visualization techniques — ranging from classical statistical charts and geospatial maps to cutting- edge interactive dashboards and AI-augmented visual analytics — and examines how each technique can be leveraged to extract actionable insights from complex datasets.
We explore the theoretical underpinnings of visual perception and cognitive load theory as they relate to visualization design, and provide a structured taxonomy of visualization types aligned to specific analytical goals. We discuss best practices for dashboard design, accessibility, color theory, and narrative data storytelling. Furthermore, we analyze emerging trends including real-time streaming visualizations, augmented reality (AR) interfaces, and the role of large language models (LLMs) in automating insight generation from visual data.
We explore the theoretical underpinnings of visual perception and cognitive load theory as they relate to visualization design, and provide a structured taxonomy of visualization types aligned to specific analytical goals. We discuss best practices for dashboard design, accessibility, color theory, and narrative data storytelling. Furthermore, we analyze emerging trends including real-time streaming visualizations, augmented reality (AR) interfaces, and the role of large language models (LLMs) in automating insight generation from visual data.
Abstract
NeoAI: Socializing Reinvented with AI
Rampal Gupta, Ms. Deepika Pandey
DOI: 10.17148/IJARCCE.2026.154296
Abstract: This research presents NeoAI, an AI-powered social media application designed to improve user experience through personalized content delivery. Unlike traditional platforms that use basic or chronological feeds, NeoAI leverages machine learning to analyze user interactions such as likes, comments, and watch time. It uses collaborative filtering on a user-item interaction matrix to recommend relevant content. The system is built with a full-stack architecture including React.js, Node.js, Express.js, MongoDB, and a FastAPI-based ML service. In addition to recommendations, the platform supports features like posts, reels, stories, messaging, and notifications. By integrating AI, NeoAI enhances engagement, content discovery, and overall usability, making social media more intelligent and user-centric.
Keywords: Social Media, Recommendation System, Collaborative Filtering, Machine Learning, React.js, Node.js, MongoDB, FastAPI.
Keywords: Social Media, Recommendation System, Collaborative Filtering, Machine Learning, React.js, Node.js, MongoDB, FastAPI.
Abstract
AI-Powered Digital Justice System: A Proposed Smart Platform for Complaint Filing and Tracking – A Review
Chilukuri Venugopal, B N Gowtham, Basava Prabhu, Chandrasekhar H, Asst. Prof. Nagateja P
DOI: 10.17148/IJARCCE.2026.154297
Abstract: Reporting crimes to police remains a tedious and inefficient experience across many jurisdictions. Citizens must physically visit police stations, complete handwritten forms, and submit complaints without receiving updates about case progress. This research proposes developing an AI-Enhanced Smart Platform for Complaint Management to address these challenges. The proposed system enables citizens to file complaints through three methods: written text, audio recordings, or video submissions. After receiving a complaint, the platform employs Natural Language Processing (NLP) to analyze content and recommend applicable Indian Penal Code (IPC) or IT Act provisions. Citizens can upload supporting evidence, monitor complaint status continuously, and interact with assigned officers via the platform. For law enforcement personnel, a specialized dashboard facilitates efficient complaint management and response. This research examines current systems, highlights their limitations, and establishes the necessity for the proposed platform. While the system remains under development, this paper presents detailed design specifications and objectives.
Keywords: Smart Complaint Platform, Artificial Intelligence, Natural Language Processing, Complaint Classification, IPC Section Mapping, Digital Policing, E-Governance Integration
Keywords: Smart Complaint Platform, Artificial Intelligence, Natural Language Processing, Complaint Classification, IPC Section Mapping, Digital Policing, E-Governance Integration
Abstract
A Comprehensive Study on Privacy-Preserving Machine Learning (PPML) in Modern AI Systems
Naveen Kumar, Shiva Kumar, Paras Kaushik, Ms. Usha Kumari, Mr. Satish Kumar Soni, Mr. Uruj Jaleel
DOI: 10.17148/IJARCCE.2026.154298
Abstract: We are living in a time when AI is no longer a futuristic concept but a daily reality. From health apps that monitor our vitals to banking systems that flag fraud, machine learning is making decisions that directly affect people's lives. But this progress comes with a serious trade-off: to make AI smarter, we feed it enormous amounts of personal data. This raises a question that researchers and policymakers are urgently trying to answer - how do we build intelligent systems without putting people's privacy at risk? Privacy-Preserving Machine Learning (PPML) is the answer the research community has been working toward. It is a growing subfield of AI that explores how machine learning models can be trained and deployed without ever needing access to raw personal data. This paper takes a comprehensive look at PPML - what it is, how it works, where it is being used, and what still stands in the way of its widespread adoption. We cover the four main privacy-enhancing techniques: Differential Privacy, Federated Learning, Homomorphic Encryption, and Secure Multi-Party Computation. We support our analysis with comparative diagrams and performance charts, and we discuss both the progress made between 2021 and 2026 and the challenges that researchers are still working to solve. Our goal is to give readers - whether students, engineers, or policy professionals - a clear and honest picture of where PPML stands today and where it needs to go.
Keywords: Privacy-Preserving Machine Learning, Differential Privacy, Federated Learning, Homomorphic Encryption, SMPC, Data Security, AI Ethics, GDPR, Deep Learning.
Keywords: Privacy-Preserving Machine Learning, Differential Privacy, Federated Learning, Homomorphic Encryption, SMPC, Data Security, AI Ethics, GDPR, Deep Learning.
Abstract
SMART EXPENSE FINANCE REPORTING USING AI
Gufran Siddiqui, Mohammad Arif, Saurabh Mishra, MD Arif, Mr. Bibhuti Kumar
DOI: 10.17148/IJARCCE.2026.154299
Abstract: A new study introduces an intelligent tool for managing money reports with help from artificial intelligence, created on the MERN platform - using MongoDB, Express.js, React.js, and Node.js. Instead of typing everything in by hand, the software tracks costs automatically, sorts them, and then builds summaries thanks to learning-based code. Because it runs without constant oversight, mistakes happen less often while workers save time and see up-to-the- minute data across clear display panels. Hidden math systems spot unusual buys, learn how users spend over weeks, yet also suggest custom saving plans based on past behavior. Out of reach? Not anymore - MERN handles growth without breaking stride. Smooth interactions come alive when parts work together, quietly boosting how people and teams manage money online. Decisions gain clarity through live insights, weaving tighter oversight into everyday actions across digital spaces. MERN Stack builds the base of modern web apps, using MongoDB, Express.js, and Node.js, React together. A system like that powers smart finance tools when tied to artificial intelligence. Instead of manual entries, automation handles routine tasks across expense management workflows. Reports form faster because financial reporting leans on real-time data flows now. With machine learning, patterns emerge from past numbers to guide future choices quietly. Data analytics turns raw figures into useful insights without extra effort. One such setup creates a smart finance system working behind the scenes daily.
Abstract
SafeRoute AI: A Comprehensive Review of Safety-Aware Intelligent Navigation Systems
E Harsha, K P Sai Pravallika, Mahima Swaroopa C K, Deekshitha B, Muhibur Rahman T R
DOI: 10.17148/IJARCCE.2026.154300
Abstract: Modern navigation systems are predominantly designed to optimize travel efficiency—minimizing distance, time, or fuel consumption—while treating personal safety as a secondary consideration. This gap is particularly significant in dense urban environments where crime rates and environmental hazards are geographically distributed in ways that conventional routing algorithms cannot account for. SafeRoute AI proposes an integrated architecture that embeds a machine learning risk prediction layer directly into the navigation decision pipeline. The system draws from structured crime statistics, geographic coordinates, and temporal signals to generate per-location risk estimates, subsequently applying a parameterized cost function that balances spatial distance against predicted danger. Route computation employs a modified A* algorithm in which a beta-weighted safety cost (β = 0.7) takes precedence over an alpha-weighted distance term (α = 0.3), ensuring that computed paths prioritize safety over raw travel efficiency. This review surveys existing literature on intelligent navigation and safety-aware systems, proposes a four-tier taxonomy classifying systems by depth of safety integration, analyzes critical research gaps in the current body of work, and outlines the complete SafeRoute AI system architecture. The analysis demonstrates that safety-aware navigation constitutes an underdeveloped yet tractable engineering challenge well- suited to current machine learning tooling, with significant potential to improve quality of life for users navigating high-risk urban environments.
Keywords: Safety-Aware Navigation; Risk Prediction; XGBoost; A* Algorithm; Crime Mapping; Intelligent Routing; Emergency Alert; Route Optimization; Machine Learning; Geographic Information Systems
Keywords: Safety-Aware Navigation; Risk Prediction; XGBoost; A* Algorithm; Crime Mapping; Intelligent Routing; Emergency Alert; Route Optimization; Machine Learning; Geographic Information Systems
Abstract
Zero Hunger: Smart Food Distribution Platform
Sharana Kumar K, Ajay Kumar Gouda S, Vaishnavi Y, MD.Akif Bari Shaik, Muhibur Rahman T.R, Anita Patil
DOI: 10.17148/IJARCCE.2026.154301
Abstract: The Zero Hunger – Food Sharing System is a web-based platform designed to reduce food wastage and support equitable food distribution by connecting surplus food donors with individuals in need. Large amounts of edible food are wasted daily due to the absence of an organized sharing mechanism, while many people struggle to access basic meals. This system addresses the gap by enabling donors to upload food details such as quantity, type, and location, and allowing receivers to easily search and request available food through a user-friendly interface.
The platform incorporates real-time listings, location-based search, secure authentication, and donation status tracking to ensure efficient and transparent food redistribution. By utilizing modern web technologies, the system improves coordination between users and minimizes food loss. Overall, the solution contributes to social welfare and aligns with the United Nations Sustainable Development Goal (SDG 2) – Zero Hunger by promoting sustainable and responsible food-sharing practices.
The platform incorporates real-time listings, location-based search, secure authentication, and donation status tracking to ensure efficient and transparent food redistribution. By utilizing modern web technologies, the system improves coordination between users and minimizes food loss. Overall, the solution contributes to social welfare and aligns with the United Nations Sustainable Development Goal (SDG 2) – Zero Hunger by promoting sustainable and responsible food-sharing practices.
Abstract
Can everything be Solar Powered? – A Comprehensive Analysis
B.md. Saif, Prashanth Patil, Rahul Naik, Harshavardhan S, Dr. Muhibur Rahaman T.R
DOI: 10.17148/IJARCCE.2026.154302
Abstract: Solar energy has emerged as a highly promising renewable energy source to address the increasing global demand for electricity and the environmental issues caused by fossil fuels. With advancements in photovoltaic technology and growing investments in sustainable infrastructure, solar power is being widely adopted across residential, industrial, and transportation sectors. This paper investigates whether it is feasible to power all systems entirely using solar energy by analyzing its potential, applications, and limitations. It examines key challenges such as energy intermittency, storage constraints, high initial costs, and geographical variability that affect consistent energy generation. The study also considers the role of modern solutions like energy storage systems and smart grids in overcoming these limitations. The analysis indicates that although solar energy can meet a significant portion of global energy needs, complete reliance on solar power alone is currently impractical. Therefore, a hybrid energy approach integrating solar with other renewable sources is identified as the most effective and sustainable path forward.
Keywords: Solar Energy, Renewable Energy, Sustainability, Photovoltaic Systems, Energy Storage
Keywords: Solar Energy, Renewable Energy, Sustainability, Photovoltaic Systems, Energy Storage
Abstract
EXPENSE MATE
Shubham, Aniket Gupta, Manish Patel, Abhishek Yadav, Mrs. Prachi Yadav
DOI: 10.17148/IJARCCE.2026.154303
Abstract: Managing personal and organizational finances has long been a tedious and error-prone task. This paper presents Expense Mate — a smart expense and financial reporting system that uses Artificial Intelligence alongside the MERN stack (MongoDB, Express.js, React.js, and Node.js) to make this process considerably easier and more reliable. Rather than relying on manual data entry, the system automatically tracks, categorizes, and generates reports for expenses using machine learning algorithms. It also offers real-time financial dashboards, detects unusual spending patterns, and gives users personalized budgeting advice. Because it is built on the MERN architecture, the system scales well and delivers a smooth user experience. The core idea behind this work is simple: by combining modern web technologies with AI, individuals and organizations can gain genuine control over their finances, make better decisions, and reduce the inefficiencies that come from doing things by hand.
Keyword: MERN Stack, Artificial Intelligence, Expense Management, Financial Reporting, Automation, Data Analytics, Machine Learning, Smart Finance System.
Keyword: MERN Stack, Artificial Intelligence, Expense Management, Financial Reporting, Automation, Data Analytics, Machine Learning, Smart Finance System.
Abstract
Smart Healthcare System for Automated Disease Detection and Cure Recommendation through Medical Imaging
Y.G. Sanjana, Shravan Kumar. V, T. Manikanta, Vaishnavi. J, Anita S Patil
DOI: 10.17148/IJARCCE.2026.154304
Abstract: The rapid advancement of medical imaging and artificial intelligence has enabled the development of smart healthcare systems for accurate and timely disease detection. This paper proposes an intelligent healthcare framework that utilizes deep learning techniques to automatically analyze medical images such as X-rays, MRI, and CT scans for early diagnosis of diseases. The system employs convolutional neural networks (CNNs) for feature extraction and classification, ensuring high accuracy and efficiency in identifying abnormalities. In addition to disease detection, the proposed model integrates a recommendation module that suggests suitable treatment options and preventive measures based on the diagnosed condition. The system is designed to assist healthcare professionals by reducing diagnostic errors, minimizing workload, and improving decision-making processes. Furthermore, the integration of cloud-based storage ensures secure data management and easy accessibility for both patients and medical practitioners. Experimental results demonstrate that the system achieves reliable performance with faster processing time compared to traditional diagnostic methods. Overall, this smart healthcare solution enhances early diagnosis, supports personalized treatment, and contributes to improved patient outcomes, especially in resource-limited environments.
Keywords: Predictive Analytics, Computer-Aided Diagnosis (CAD), Healthcare Automation, Medical Image Analysis, Early Detection Systems, Personalized Medicine, Cloud-Based Healthcare, Big Data in Healthcare, Diagnostic Accuracy, Pattern Recognition, Neural Networks, Clinical Decision Making, Health Informatics, Remote Patient Monitoring, AI- Based Healthcare
Keywords: Predictive Analytics, Computer-Aided Diagnosis (CAD), Healthcare Automation, Medical Image Analysis, Early Detection Systems, Personalized Medicine, Cloud-Based Healthcare, Big Data in Healthcare, Diagnostic Accuracy, Pattern Recognition, Neural Networks, Clinical Decision Making, Health Informatics, Remote Patient Monitoring, AI- Based Healthcare
Abstract
ArogyaAI: An Integrated LLM-Powered Healthcare Intelligence System
Chetan, Harshit Kumar, Atul Tyagi, Anshul Rastogi, Divyanshi Muvaniya, Dr. Brijesh Kr. Gupta
DOI: 10.17148/IJARCCE.2026.154305
Abstract: Healthcare systems continue to struggle with fragmented patient records, poor AI transparency, and limited telemedicine integration. ArogyaAI is a web-based platform built using Django and SQLite that addresses these issues in one place. It uses the LLaMA 3.3 70B language model through the Groq API to provide symptom guidance and automatically summarize medical reports. Integrated video consultations run directly in the browser via Jitsi Meet, and access is controlled based on user role—patient, doctor, or administrator. Testing shows the system handles multilingual conversations (English, Hindi, Hinglish), simplifies appointment workflows, and reduces the difficulty of reading clinical documents. The design demonstrates how modern web tools and large language models can work together to improve healthcare access.
Keywords: Artificial Intelligence, Electronic Health Records, Generative AI, Healthcare Web Application, LLaMA, Telemedicine.
Keywords: Artificial Intelligence, Electronic Health Records, Generative AI, Healthcare Web Application, LLaMA, Telemedicine.
Abstract
Customer Churn Analysis Using Python and Data Analytics for Telecom Industry
Kartik Bhargav, Khushi, Lalit Kumar, Mayank Bhardwaj, Dr. Brijesh Kumar Gupta
DOI: 10.17148/IJARCCE.2026.154306
Abstract: The global telecommunication sector is currently navigating an era of rapid transformation driven by intense market competition. This paper work explores the intersection of data science and business intelligence, examining how analytical methodologies can be leveraged to mitigate customer churn—the rate at which subscribers terminate service relationships. Through a comprehensive analysis of telco datasets, this study demonstrates that identifying "at-risk" behaviour before departure is critical for revenue stability, as the cost of acquiring new customers is 5 to 6 times higher than retaining existing ones. We present a data-driven framework for churn assessment that incorporates a specialized Injection Layer for automated log collection and data handling, an Analytical Layer utilizing an engine for deep-dive SQL querying, and a Visualization Layer for interactive reporting via Power BI. The findings reveal that significant predictors of churn include high monthly charges and short-term contract types. Furthermore, the implementation of an ML-based optimization framework achieved an accuracy score of 0.843 using an Extra Trees Classifier, proving that predictive modelling can achieve 40-60% reductions in attrition when integrated into core management strategies. This study contributes to the field by providing actionable insights for organizations seeking to optimize resource allocation while maintaining long-term organizational sustainability.
Keywords: Analytical Engine, Customer Churn Management, Injection Layer, Predictive Modelling.
Keywords: Analytical Engine, Customer Churn Management, Injection Layer, Predictive Modelling.
Abstract
Survey-Based Research in Software Engineering
Kavya K, Layashree L, Raziya M, S Vaishnavi, Dr. Muhibur Rahman T.R
DOI: 10.17148/IJARCCE.2026.154307
Abstract: Survey-based research in software engineering is an empirical method used to collect data from developers and organizations to understand current trends, challenges, and practices. This paper discusses survey design, sampling techniques, and data collection methods to ensure reliable and valid results. It highlights the role of statistical analysis in interpreting data and deriving meaningful insights. The study concludes that survey research supports improved software development processes and evidence-based decision-making.
Keywords: Include at least 4 keywords or phrases.
Keywords: Include at least 4 keywords or phrases.
Abstract
FinGenie: Design and Implementation of an AI-Assisted Financial Literacy Platform for Adolescents
Sowmya Prabhu, Soumya Sree T., Roopa Sai T., Venkat Ravi, Muhibur Rahman T.R, Anita Patil
DOI: 10.17148/IJARCCE.2026.154308
Abstract: Financial illiteracy among teenagers continues to be a major challenge in education systems around the world. A large number of young people step into adult financial life with little to no understanding of how money actually works. This paper presents the design, development, and real-world evaluation of FinGenie — a locally hosted, AI-powered web application built specifically to help learners between the ages of 13 and 19 develop meaningful financial skills. Rather than relying on third-party cloud services, FinGenie uses on-device language models to power a Retrieval-Augmented Generation (RAG) chatbot, protecting user privacy while still delivering intelligent, personalized responses. The platform also includes structured learning modules, gamified elements like quests and badges, and hands-on tools such as a digital wallet and goal tracker. Built on a React–TypeScript frontend and an Express-based RESTful backend with the Ollama inference runtime, the system was tested with 150 adolescent users over an eight-week period. Results showed a 26- percentage-point increase in financial knowledge scores, an 84% retention rate at the 30-day mark, and an overall satisfaction rating of 87%. These outcomes suggest that well-designed educational technology — when combined with responsible AI use and strong pedagogical foundations — can produce real, lasting improvements in financial capability for teenage users.
Keywords: Financial literacy, adolescent education, Retrieval-Augmented Generation, local AI, gamification, React, Node.js, Ollama, educational technology, privacy-first design
Keywords: Financial literacy, adolescent education, Retrieval-Augmented Generation, local AI, gamification, React, Node.js, Ollama, educational technology, privacy-first design
Abstract
Classification and Analysis of Disease over Symptoms using AI
Peetambar, Amir Ansari, Samrat Kartikey Maurya, Mr. Dileep Kumar Gupta
DOI: 10.17148/IJARCCE.2026.154309
Abstract: This paper presents a smart healthcare system that integrates machine learning-based disease prediction with an online appointment booking platform. The system aims to address the problem of delayed diagnosis and limited access to medical consultation by combining predictive analytics with real-time healthcare services. Machine learning algorithms such as Support Vector Machine (SVM) and Logistic Regression are used to predict diseases like diabetes, heart disease, and Parkinson’s based on user input data.
In addition to prediction, the system provides a web-based interface where patients can easily browse doctors and book appointments. The platform includes role-based dashboards for patients, doctors, and administrators, enabling efficient management of appointments, patient records, and doctor availability. The integration of prediction and booking ensures that users can take immediate action after receiving health insights.
The proposed system improves early detection, reduces manual effort, and enhances accessibility to healthcare services. It also provides a scalable and user-friendly solution that can be deployed in real-world healthcare environments. This approach demonstrates how the combination of machine learning and full-stack development can significantly improve healthcare delivery and patient outcomes.
The system is designed to be user-friendly, scalable, and efficient for real-world healthcare applications. It ensures secure data handling and smooth integration between prediction and booking modules. The approach reduces the gap between diagnosis and consultation, improves patient engagement, and helps healthcare providers manage patient flow effectively.
Keyword: Artificial Intelligence, Machine Learning, Disease Prediction, Healthcare Analytics, Diabetes Prediction, Heart Disease Detection, Parkinson’s Prediction, Predictive Modeling, Clinical Decision Support System, Streamlit Web Application, Medical Data Analysis, Early Diagnosis, Health Monitoring System, AI in Healthcare, Appointment Booking System Integration
In addition to prediction, the system provides a web-based interface where patients can easily browse doctors and book appointments. The platform includes role-based dashboards for patients, doctors, and administrators, enabling efficient management of appointments, patient records, and doctor availability. The integration of prediction and booking ensures that users can take immediate action after receiving health insights.
The proposed system improves early detection, reduces manual effort, and enhances accessibility to healthcare services. It also provides a scalable and user-friendly solution that can be deployed in real-world healthcare environments. This approach demonstrates how the combination of machine learning and full-stack development can significantly improve healthcare delivery and patient outcomes.
The system is designed to be user-friendly, scalable, and efficient for real-world healthcare applications. It ensures secure data handling and smooth integration between prediction and booking modules. The approach reduces the gap between diagnosis and consultation, improves patient engagement, and helps healthcare providers manage patient flow effectively.
Keyword: Artificial Intelligence, Machine Learning, Disease Prediction, Healthcare Analytics, Diabetes Prediction, Heart Disease Detection, Parkinson’s Prediction, Predictive Modeling, Clinical Decision Support System, Streamlit Web Application, Medical Data Analysis, Early Diagnosis, Health Monitoring System, AI in Healthcare, Appointment Booking System Integration
Abstract
Ethical and Legal Implications of Using Generative AI in Academic Research
M S Saniya Khouser, Meghana D, Meghana Y, Mounika Y R, Dr. Muhibur Rahman T.R
DOI: 10.17148/IJARCCE.2026.154310
Abstract: The rapid adoption of generative Artificial Intelligence (AI) tools in academic research has introduced significant ethical and legal challenges that demand critical examination. While these technologies enhance productivity by assisting in literature review, data analysis, and content generation, they also raise concerns regarding authorship, originality, accountability, and intellectual property rights (IPR). This study explores the ethical implications of using generative AI in research, including issues of plagiarism, bias in AI-generated content, and the erosion of academic integrity. Furthermore, it analyzes the legal uncertainties surrounding ownership of AI-generated work, copyright infringement, and the absence of clear regulatory frameworks governing AI usage in academia. Using a qualitative research methodology, this paper reviews existing literature, case studies, and policy guidelines to identify gaps in current practices. The findings highlight the need for transparent usage policies, proper attribution mechanisms, and updated IPR laws to address emerging challenges. The study concludes by proposing recommendations for responsible AI usage that balance innovation with ethical and legal compliance in academic research.
Abstract
MedRAG Nexus: An AI-Powered Health Intelligence System Using Retrieval-Augmented Generation and Agentic AI
Sandeep Tomar, Abhishek Soam, Shekhar Tomar, Tanya Chaudhary, Sandhya Kashyap, Dr. Brijesh Kr. Gupta
DOI: 10.17148/IJARCCE.2026.154311
Abstract: A significant and underappreciated challenge in modern healthcare is the communicative divide between clinical documentation and the patients those documents describe. Pathology reports, handwritten prescriptions, and physician summaries are routinely generated but rarely understood by the individuals who receive them — creating a measurable gap between information delivery and informed patient action. This paper presents MedRAG Nexus, a tri- layered AI-powered health intelligence platform designed to close this gap through a RAG-grounded conversational interface. The system processes clinical documents via a multimodal Vision-AI pipeline — employing a fine-tuned TrOCR model for handwritten prescription parsing and an EfficientNet-B7 network for dermatological anomaly classification — and anchors all clinical reasoning in a Retrieval-Augmented Generation (RAG) framework built on ChromaDB and LangChain. Patients interact with their own medical records through natural language; the chatbot retrieves verified clinical knowledge before generating every response, ensuring that answers are evidence-grounded rather than model-generated. A LangGraph-orchestrated ReAct agentic layer enables autonomous interventions — including specialist appointment scheduling via Google Calendar and patient notification via the Twilio WhatsApp API — when document analysis identifies clinically significant findings. The architecture directly addresses two dominant failure modes of current health AI: inaccessible medical documentation and factual hallucination in general-purpose large language models (LLMs). Experimental design targets a prescription parsing accuracy exceeding 92%, a hallucination rate approaching zero through retriever-grounding, and a demonstrable reduction in patient time-to-care. This paper details the full system architecture, data flow, model selection rationale, ethical safeguards, and a structured validation protocol suitable for clinical evaluation.
Keywords: Retrieval-Augmented Generation (RAG); Large Language Models; Clinical Document Understanding; TrOCR; EfficientNet; LangGraph; Agentic AI; Clinical NLP; ChromaDB; Health Informatics; ReAct Framework; Medical AI; Prescription Parsing; Patient Health Literacy.
Keywords: Retrieval-Augmented Generation (RAG); Large Language Models; Clinical Document Understanding; TrOCR; EfficientNet; LangGraph; Agentic AI; Clinical NLP; ChromaDB; Health Informatics; ReAct Framework; Medical AI; Prescription Parsing; Patient Health Literacy.
Abstract
Full-Stack Hospital Management System
Shivraj, Varun Sirohi, Saurabh Kumar Nirwan, Shivansh Chaudhary, Tanya Jain, Dr.Uruj Jaleel, Dr. Satish Kumar Soni
DOI: 10.17148/IJARCCE.2026.154312
Abstract: This paper presents MediCare Pro, a comprehensive full-stack Hospital Management System (HMS) developed using FastAPI, SQLAlchemy, and modern web technologies. The system addresses the growing need for digitised, integrated healthcare administration by providing a unified platform covering patient registration and management, doctor scheduling, appointment booking, billing with UPI payment integration, medical records, inventory control, staff management, and real-time analytics. A distinguishing feature of MediCare Pro is its incorporation of biometric face recognition login using face-api.js, eliminating the need for manual credential entry. The backend exposes 30+ RESTful API endpoints, while the frontend is a Single-Page Application (SPA) delivering a responsive, role-aware dashboard. Evaluation demonstrates that the system achieves complete CRUD coverage across eight data entities, real- time KPI monitoring through 14 computed statistics, and seamless UPI-based payment flows. The paper details the system architecture, data model, module design, security considerations, and future enhancement roadmap.
Keywords: Hospital Management System, FastAPI, SQLAlchemy, Face Recognition, UPI Payment, RESTful API, Single-Page Application, Healthcare Digitisation, SQLite, Python
Keywords: Hospital Management System, FastAPI, SQLAlchemy, Face Recognition, UPI Payment, RESTful API, Single-Page Application, Healthcare Digitisation, SQLite, Python
Abstract
AI Based Food Donation and Distribution System
B Bhargavi, Chandana B A, Kasireddy Rithu Reddy, V R Kalyani, Anita Patil
DOI: 10.17148/IJARCCE.2026.154313
Abstract: There are major problems faced in the world like Food wastage and Hunger. Sometimes there is a large quantity of food wastage, even though people are facing issue of having proper meals. With the development of many technologies like Artificial Intelligence (AI), Machine Learning (ML), and the Internet of Things (IoT), they developed smarter food donation systems. This paper reviews suggests different research works related to food donation and distribution system. The smarter food donation system focus on features such as user interaction, food allocation, delivery management, and prediction techniques. The research also focuses on the drawbacks of ongoing systems and proposes the need for a more advanced and integrated solution.
Keywords: Food Donation, Food Distribution, Artificial Intelligence, Machine Learning, IOT, Smart Systems.
Keywords: Food Donation, Food Distribution, Artificial Intelligence, Machine Learning, IOT, Smart Systems.
Abstract
AI-Based Mental Health Chatbot using LangChain and LangGraph
Sachin saini, Yashika Rathi, Yash Kumar, Shweta Rani, Dr Brijesh kr. Gupta
DOI: 10.17148/IJARCCE.2026.154314
Abstract: Mental health disorders such as depression, anxiety, and stress are increasing at an alarming rate worldwide, while access to professional mental healthcare remains limited due to cost, stigma, and resource constraints. Artificial Intelligence (AI)-based mental health chatbots have emerged as scalable and accessible solutions to address this gap. This paper presents a comprehensive study and proposed architecture for an AI-based mental health chatbot using LangChain and LangGraph frameworks. LangChain enables modular, context-aware conversational capabilities with memory and knowledge integration, while LangGraph introduces multi-agent workflows, state management, and human-in-the-loop control for improved safety and decision-making. The proposed system integrates Natural Language Processing (NLP), Cognitive Behavioral Therapy (CBT), sentiment analysis, and Retrieval-Augmented Generation (RAG) to deliver personalized and context-aware responses. The paper also reviews existing literature, discusses system architecture, methodology, evaluation techniques, and highlights challenges such as ethical concerns and safety risks. The results from existing studies demonstrate that AI chatbots significantly improve mental well-being, although further clinical validation is required.
Abstract
Enhanced Distributed Flood Detection and Alert System Using Deep Learning
NageswaraRao Sirisala, Srinivasulu Sirisala, Anitha Yarava
DOI: 10.17148/IJARCCE.2026.154315
Abstract: Among natural disasters, flood has greater impact on farming, property, human lives and their money situation, which further affects the overall economy. We are seeing that AI alert systems for floods are only necessary to reduce the damage. Basically, this paper proposes the Federated Learning approach for flood prediction model. Federated Learning itself is a distributed ML technique that further reduces data transfer from flood sites to the central server. As per the system design, FL can give security and privacy to data and keep it available always. Regarding data protection, this method ensures information stays safe and accessible at all times. Moreover, in FL based flood prediction model, each location trains its individual model using local flood data, and this trained model is further shared with the server itself. The server combines all local models to prepare a global model further. This process itself creates a unified model from individual contributions. We are seeing this method can avoid network delays only, so the global model can take quick decisions on floods. As per this paper, the server combines local models to make a global model regarding five- day flood warning plans for each place. We are seeing that a local model is trained using regional conditions to give predictions on possible flood areas and their maximum water levels only. The data set actually has flood history information from five rivers between 2015 to 2021, which is used for training the model. Basically, it includes four features - rainfall overflow, snow melting rate, water movement dynamics, and current river flow, which are the important factors for water analysis. Basically, the results show flood prediction data collected from 2010 to 2015 for the selected area using the proposed method with 84% accuracy. The model is further improved by adding a Convolutional 2D Neural Network (CNN2D) itself. The improved version of our proposed method can enhance flood predictions with 90% accuracy. Moreover, this approach provides reliable results for better flood forecasting.
Keywords: Feedforward Neural Network (FFNN), Random Forest, Decision Tree, Gradient Boost, Convolutional Neural Network (CNN).
Keywords: Feedforward Neural Network (FFNN), Random Forest, Decision Tree, Gradient Boost, Convolutional Neural Network (CNN).
Abstract
Deep Learning Framework for Prior Identification of Threats in Social Media Interactions
NageswaraRao Sirisala, Srinivasulu Sirisala, Anitha Yarava
DOI: 10.17148/IJARCCE.2026.154316
Abstract: Social Media Interactions are digital platforms that enable users to create profiles, connect with others, and interact through various forms of communication. These platforms facilitate social interaction, content sharing, and collaboration across geographical boundaries. Threats in Social media interactions refer to risks and malicious activities that exploit vulnerabilities in online platforms, posing harm to users, data, and digital ecosystems. These threats can impact personal security, privacy, and overall platform trustworthiness. The Deep Learning-Based Framework for Prior Identification of Threats (DLPIT) is designed to proactively detect and mitigate harmful content in social media interactions, such as hate speech, cyberbullying, and violent language, before they escalate. By leveraging a Recurrent Neural Network (RNN), which is particularly effective for sequential data processing, the system analyzes Twitter language and classifies information as either harmful or non-threatening. The framework is trained on a preprocessed labeled Twitter dataset that incorporates both textual and behavioral data, ensuring comprehensive threat detection. The RNN's ability to capture contextual relationships and temporal dependencies enables DLPIT to monitor social media platforms in real time with high efficiency. Furthermore, the framework enhances detection accuracy by integrating social network interactions and user engagement patterns, which help in identifying the potential influence and reach of harmful content. To quantify the severity of a detected threat, the system calculates a Threat Level Score (TLS) based on multiple factors, including the intensity and frequency of harmful words, user history, past engagement patterns, and the influence of the content within the social network. A higher TLS signifies a greater risk, enabling moderators to prioritize intervention and take necessary actions accordingly. The performance of DLPIT is rigorously evaluated and compared with existing methods using F1-score, recall, accuracy, and precision.
Keywords: Recurrent Neural Network (FFNN), Threat Level Score, Explainable AI, Cyber Bulling
Keywords: Recurrent Neural Network (FFNN), Threat Level Score, Explainable AI, Cyber Bulling
Abstract
A Study on Multi-Location Ledger Using DPOS (DP46) in a FPL Hyundai Automobile Dealership
Kumar. Nandini, Dr. Kumarakannan.R
DOI: 10.17148/IJARCCE.2026.154317
Abstract: Automobile dealerships generate large volumes of financial data every day through their Sales, Service, and Parts departments. When a dealership operates across multiple locations, it becomes increasingly difficult to maintain accurate, synchronized, and up-to-date ledgers. Delays in data entry, mismatch in branch-wise reports, duplication of transactions, and manual posting errors can lead to major financial discrepancies and delays in monthly closing activities. To address these issues, Hyundai dealerships use the Dealer Parts Operations System (DPOS), a digital platform designed to automate ledger posting, stock management, billing, and financial reporting. This study focuses on evaluating how effectively DPOS supports multi-location ledger management in a dealership environment.
The research adopts a mixed-method approach using interviews with accounts staff, direct observations of DPOS operations, and analysis of ledger reports and audit data. The findings show that DPOS significantly enhances accuracy, reduces manual intervention, improves real-time data synchronization between branches, and speeds up monthly financial consolidation. However, challenges such as system slowdowns, dependency on user skill, and occasional posting errors still exist.
Keywords: DPOS, Multi-Location Ledger, Financial Management, Automobile Dealership, Ledger Automation, Monthly Accounts, Inventory and Parts Management, Digital Accounting System.
The research adopts a mixed-method approach using interviews with accounts staff, direct observations of DPOS operations, and analysis of ledger reports and audit data. The findings show that DPOS significantly enhances accuracy, reduces manual intervention, improves real-time data synchronization between branches, and speeds up monthly financial consolidation. However, challenges such as system slowdowns, dependency on user skill, and occasional posting errors still exist.
Keywords: DPOS, Multi-Location Ledger, Financial Management, Automobile Dealership, Ledger Automation, Monthly Accounts, Inventory and Parts Management, Digital Accounting System.
Abstract
A STUDY ON THE ANALYSIS OF POST-PURCHASE EXPERIENCE USING POWER BI WITH REFERENCE TO SOLSTROM ENERGY SOLUTION PVT LTD
Mr. S. Balakrishnan, Dr. K. Kokila
DOI: 10.17148/IJARCCE.2026.154318
Abstract: The transition to renewable energy increased the necessity of the organizations to improve customer satisfaction and their service quality, particularly in solar energy industry. This paper provides an analysis of the post- purchase customer experience in Solstrom Energy Solutions based on a mathematical evaluation of customer experience. The 102 customers were the sample population who participated via a structured questionnaire to gather primary data on product performance, quality of installations, transparency in communication and after sales services which are critical drivers of long-term satisfaction to the adoption of sustainable energy. Results have shown that solar system effectiveness and parts dependability were rated highly on their satisfaction ratings, and this validates high technical efficacy. The quality of installation and professionalism of service were rated positively, and this aspect added to the trust. The dashboard helped to process raw feedback to meaningful indicators that allowed making decisions faster, thereby facilitating sustainable service delivery. The paper concludes that by integrating business analytics, the strategies can be made much more customer-centric, more willing to implement renewable energy solutions, and make long-term sustainability objectives.
Keywords: Solar Energy Systems, Data-Driven Decision Making, Sustainable Service Practices, Customer Feedback Analysis.
Keywords: Solar Energy Systems, Data-Driven Decision Making, Sustainable Service Practices, Customer Feedback Analysis.
Abstract
A STUDY ON IMPACT OF SOCIAL MEDIA MARKETING ON CONSUMER BUYING BEHAVIOUR - DIGITAL MARKETING SECTOR, CHENNAI.
Ms. K. Thrisha, Dr. K. Kokila
DOI: 10.17148/IJARCCE.2026.154319
Abstract: Nowadays, businesses rely heavily on social media to shape how people decide what to buy - it's part of everyday digital outreach. Looking closely, this work explores how those online efforts sway shopper choices. Information came straight from users answering questions online, mainly folks active on Instagram, Facebook, or YouTube. A clear structure guided the approach, focusing on individuals easy to reach through shared networks. To spot patterns, methods like correlation and regression ran inside IBM SPSS Statistics software. Numbers helped show just how strong the links are between platform activity and actual purchases.
What researchers found shows social media features - like ads, special offers, or feedback from users - shape how people decide what to buy. Because of these tools, companies can grow recognition for their name, gain confidence from buyers, while sparking real conversations. In the end, smart use of social networks shifts how customers act, becoming key within today’s online promotion Landscape.
Keywords: Brand Engagement, Consumer reviews, Social media marketing, Online social network services, Brand awareness, Purchase intent, Internet advertising.
What researchers found shows social media features - like ads, special offers, or feedback from users - shape how people decide what to buy. Because of these tools, companies can grow recognition for their name, gain confidence from buyers, while sparking real conversations. In the end, smart use of social networks shifts how customers act, becoming key within today’s online promotion Landscape.
Keywords: Brand Engagement, Consumer reviews, Social media marketing, Online social network services, Brand awareness, Purchase intent, Internet advertising.