← Back to Archives
Keywords: Portable Drilling Machine, Remote Control, Wireless Operation, DC Motor, Automation, Microcontroller, Industrial Safety.
VOLUME 15, ISSUE 5, MAY 2026
Publication in progress..
Data-Driven Analysis of Mortality and Survival Outcomes in Patients with Liver Cirrhosis: A Global Healthcare Analytics Study
Mehdi Mostofi, Prasanna Poreddy, Jayasri Katragadda, Nadeem Basha Syed, Sunandamala Bodaballa
An Effective Heart Disease Detection and Severity Level Classification Model Using Machine Learning and Hyperparameter Optimization Methods
Dr. Ganesh.G. Taware, Miss.P.D. Nale, Kaveri Kalbhor, Gaurav Karande, Swaraj Navale, Yogesh Gaikwad
Smart Hybrid Bus Tracking System with Mobile Sensing and Real-Time Updates
Anil Kumar T R, Harshavardhan Pattar, Indusri J, Joshi Chetan, Dr. Yeresime Suresh, Dr. Anita Patil
Smart Health Tracker with Doctor Consultation
Satyam Sahu, Jitendra Gupta, Suraj Verma, Mr. Kumar Bibhuti
Smart Food Donation System: A Web-Based Approach for Efficient Food Redistribution–A Review
Akshay, Prashanth K, K.Shiva darshan, Kishan kumar B.V, Pavan kumar P, Anitha Patil P
AI-Based Video surveillance System
Afsa Saboo, B Sai Dikshitha, C Mohammad Athiq, K Sudeep Gouda, Nagateja P, Anita Patil
Smart AI-Integrated Exam Security Gate
A Sunitha, B Lavanya, B Pallavi, Manisha Patel, Anita Patil
SMART RAILWAY TRACK HEALTH MONOTORING SYSTEM by IOT
Dr. C.N. Deshmukh, Mayur S. Salode, Aryan R. Kale, Minal G. Lonkar, Kshitij M. Thotange
Farm2Door: A Smart Digital Platform for Farmer-to-Customer Agricultural Marketplace-A Review
K Naveen, K Shyam, Kothapalli Vamshi, Kothapalli Vardhan Babu, Dr. Anita Patil, Asst.Prof. Shashikantha Raddi
CLEARBON: A Blockchain-Based Dual Market System for Real-Time Carbon Credit Verification and Trading
Sohail, Sooraj B, Suma Shree S, Swetha Rani H S, Anita Patil
EVOLUTION OF AI TECHNOLOGIES IN THE EDUCATIONAL SECTOR
Yashvardhan Chahal, Yash Goswami, Tushar Mahaur, Yash Jain, Dr. K.S. Mishra
Talent Track Workforce Employment Solutions Platform
Sathiyaa Balajee R, Mrs. Saranya R
SMART EXAM SEAT ALLOCATION SYSTEM USING ANDROID APPLICATION
Riya Bhoir, Bhumika Beldar, Kanishka Chindarkar, Prof. Smita Chunamari
“Smart AI-Based Student Attendance System with Monthly Analytics”
Harsh Sharma, Miss. Taniya Jain, Dr. Uruj Jaleel, Dr. Satish Kumar Soni
AI-Driven Personalized Education Platform: Design, Architecture, and Implementation
Sajid Raza, Vishal Kumar Sah, Rishabh Tiwari, Saif Siddiqui, Dr. Sandeep Kumar Dubey
Artificial Intelligence in Cybersecurity: A Comprehensive Survey on AI-Driven Insider Threat Detection
Deepak Kumar G., Deevaraj M., H Pramodh, Lakshmi Narayana, Dr. Muhibur Rahman T R
A Comprehensive Study on Data Storage Security Issues and Services in Cloud Computing
Sujal Satish Godse, Bhushan Sunil Matsagar, Kirti Dinkar More
SMART ROAD ACCIDENT DETECTION AND RESCUE SYSTEM
JEEVITHA R, B SAHANA REDDY, G TEJASHWINI, SHREYA B R
MultiCart- Ai Based Multi Vendor Cart Platform
Kamlesh Kumar Pal, Abhishek Gupta, Harshit Mall, Mr. Deepak Kumar
Public Transport Tracking Systems: A Comprehensive Study of Real-Time Solutions for Small Cities
C Renuka, Divya Rajeev M, J Kavyashree, Bindu AV, Muhibur Rahman T R
CLICKET – Be There or Make It Happen With AI
Sharad Singh, Ritik Chaurasiya, Shivam, Prince Yadav, Kumar Bibhuti Bhushan Singh
AWAZE-E-JANATA- VOICE OF THE PEOPLE
Rajveer Pratap Singh, Aman Tiwari, Raj Yadav, Mr. Dileep Kumar Gupta
Smart Battlefield Helmet with Soldier Monitoring System
Dr. Pramod Sharma, Aryansh Gupta, Divya Gupta, Jayanth Bharadwaj, Sagar Mishra
AUTOFACE - Attendance Simplified Through Vision
Shrestha Gupta, Shreoshi Roy
DejaView: An AI-Powered Academic Project Integrity and Evaluation Portal Using Vector Embeddings and Large Language Models
Mrs. Trupthi Rao, Mr. Sumit Kumar Yadav, Aditya Raj Swain, Mahadev Jagtap, Pratik, Gajanan
EMPLOYMENT HIRING PLATFORM USING AI
Sanjay Kumar, Prince Pandey, Md Shahjad, Deepak Kumar
A Review of Voice-Driven Accessible Vocational Training Platforms for Persons with Visual Impairment
Bharath Reddy, Chinta Vishnu Vardhan, H M Sharanabasava, K Ravi Kumar, Dr. Muhibur Rahman
NeuroSecure: A Comprehensive Survey on Deep Learning Approaches for Cyber Defense and Intrusion Detection
P. Anirudh, B. Adarsh Reddy, G. Raghavendra, K. Naveen Kumar, Dr. Muhibur Rahman T R
Case Study on Experimental Setup Design in Robotics Research: A Comprehensive Review of Future Directions
M A Kaif, Md Akif Bari, Md Muntaseeb, Punith Raj, Dr. Muhibur Rahaman T.R
A Review on Machine Learning Techniques for Crop Yield Prediction
A Renukamma, Arathi C G, Aruna B, K Ananya, Dr. Muhibur Rahman T.R
AI-Based Real-Time Traffic Congestion Prediction and Signal Optimization System
B Deepika, Basavarajeshwari, D R Pallavi, D Suhasini, Dr. Muhibur Rahman T.R
Crop Disease Detection Using AI: A Comprehensive Survey
G Sai Supriya, G Sanjana, Harshita S S, V Indu, Muhibur Rahman T.R
Development Of Eye Disease Detection Using Deep Learning
Afrin Amin Makandar, Pranav Sanjay Kumbhar
AUTONOMOUS/TELE-OPERATED LANDMINE DETECTION ROBOT
Deepak Yadav, Kundan Gautam, Sarfaraz Ansari, Shailesh Srivastava, Roshni dubey, Prof. Prabhakar dubey
AGRITRADEHUB: A MERN Stack Based Digital Platform for Advancing India's Farmer-Merchant Direct Trade Ecosystem
Mukesh Gupta, Dheeraj Singh, Nidhi Shukla, Roshni Soni, Ms. Shreoshi Roy
CONTENT FILTER BASED MOVIE RECOMMENDATION SYSTEM USING AI AND ML
Asharf Khan, Alok Kumar Singh, Vishal Singh, Akrar Khan, Pankaj Kumar Gupta
Alzheimer’s Disease Detection Using AI
Divyanshi Kashyap, Divya pandey, Mansi Pandey, Divyanshi Yadav, Dr. Nikhat Akhtar
Alzheimer’s Disease Detection Using AI
Divyanshi Kashyap, Divya pandey, Mansi Pandey, Divyanshi Yadav, Dr. Nikhat Akhtar
Student Health and Stress Prediction System using Machine Learning
A Revanth, G J Sachidananda, G Mahesh Gouda, Dr. Muhibur Rahman T.R
UNIEVENTS : A COLLEGE EVENT HANDLER USING AI
Laxmi Kanaujiya, Tanya Prajapati, Samriddhi Malhotra, Ranjana Kumari Chirania, Ms Prachi Yadav
AI-Based Smart Home Intrusion Detection & Alert System Using Behavior Analysis and Face Recognition
Mallappa H, Yashwanth T M, Naveendra Reddy, Ameer S, Asst. Prof. Rajashekar Reddy P, Dr. Anita Patil
Generative AI: A Comprehensive Survey on Transformer-Based Models
J Hemanth, Harsha B, Ashish Kumar, Anurag N, Dr. Muhibur Rahman T.R
AI BNB: AI-BASED TRAVEL LISTING AND BOOKING PLATFORM
Gaurav Yadav, Tanveer Alam, Balbir Singh, Rishabh Shukla, Ms. Deepika Pandey
Smart Career Platform: An AI-Driven Personalised Career Guidance System for Multi-Domain Professionals Using MERN Stack and Groq Large Language Models
Purshottam Mishra, Awadhesh Kumar, Mr. Dileep Kumar Gupta
A STUDY OF FINANCIAL STATEMENT ANALYSIS OF HDFC BANK
Rushikesh Prakash Kapate, Prasad Shivajirao Kadam, Dr. Pradyuman Shastri
ZIGBEE BASED SMART HELMET FOR MINERS
Payal Survase, Pratik Ghortale, Ashwini Adasre
Blockchain-Based Secure Voting System
Palak S Sachar, Sania Shaikh, Muktha Reddy, Dr. Muhibur Rehman T.R
Smart Hospital Management System
Rakesh Yadav, Arvind Yadav, Bhanu Chandra
AFFECTIVE COMPUTING TECHNIQUES FOR HUMAN–MACHINE INTERACTION
Rabiya Fathima, C S Swetha
SafeHer: A Proactive AI-Driven Safety System for Women Using Contextual Risk Assessment and Real-Time Monitoring
Rohit Kumar Yadav, Aditya Upadhyay, Kajal Kasaudhan, Dr. Nikhat Akhtar
COVID SAFETY SYSTEM – DOOR HANDLE SANITIZER & TEMPERATURE DETECTION
Bhalghare Kajal Sanjay
AI-Powered Document Question Answering System Using Agentic RAG and Local Language Models
Varshitha D, Vidya S
Blockchain Driven Multi Tier Academic Credential Validation System
G. Priyadharshini M.E., Prakash B, Logesh P
IndianAura: A Secure and Scalable E-Commerce Platform for Certified Made-in-India Products
Aishwarya V S, Anagha I V, Bindu B, Deepika G S, Veeramreddy Rajasekhar
A Secure Task Management System for Employee Workflow Optimization
G. Priyadharshini M.E., Priyadharshene R, Vimandhani G
AI-Enabled Eye Screening Tool for Early Detection of Common Eye Diseases
G. Priyadharshini M.E., Balananthakumar B, Jancy M
Intelligent Taste Prediction Engine Using Scroll, Hover & View-Time Patterns in E-Commerce
G. Priyadharshini M.E., Roshani C, Pradeesh T
Seasonal Crop Price Prediction using AI/ML
Aniket Sawant, Harsh Dudhal, Tejas Jadhav, Sarthak Kakade, Mrs. Dhage T.S.
HELMET PROTECTION DETECTION SYSTEM USING YOLOV8 FOR REAL-TIME TRAFFIC SAFETY MONITORING
Ramya Patani, Dr. Darapu Uma
Phishing URL and Text Detection
R. Janaki M.E., (phd), Akalya A, Dharshini J
An Integrated IoT and Web-Based Framework for Energy and Machine Monitoring in SMEs
S Abhinaya, M.E., Akash M, Manimaran B
Detection of thyroid stages classification by convolutional neural network techniques
R. Janaki M.E.(phd), Ramya Shree V, Sweatha N
PHISHING WEBSITE DETECTION SYSTEM USING ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING
Mr. D.S. Jaybhay, Ms. Chaitrali S. Shinde, Ms. Bhakti D. Nannaware, Ms. Sakshi A. Harnawal Ms. Priyanka S. Gadhe
AI Driven IDS System in Network Security
Kartik P Shetty, Prof. Theerthashree G S
Brain Tumor detection using Artificial Intelligence
Rushikesh Todekar, Shejal Kawale, Sakshi Khankar, Mayuri Sudake, Dr. Sachin Bere, Prof. Mrs. Jagtap P.S
TimelimeX – Exploring the History using Gen AI
Dr. C N Shariff, Anjali, B Aishwarya, Bhavana Joshi, G Sahethi
Secure Digital Identity Verification Using Blockchain
K Senbagam M.E., Kishore S, Divya K
VisionGuard: Deep Learning–Based Weapon Detection Framework
K. Rajavadhani, Aswini. R. C & Vijayalakshmi. J
Review on Potential Applications of Geopolymer Concrete in Construction
Vasanth S, Likhith Kumar T, Vamshi A, Mohammed Afiz Roshan
CallMind: An AI-Powered Real-Time Voice Agent Platform Using a Pluggable Speech-to-Text, Large Language Model, and Text-to-Speech Pipeline
Sahil Makandar, Rohini Magamdum, Aditya Mali, Sudesh Kumbhar, Radheshyam Sah, Rajendra Hiremath
Smart Pothole Detection and Rapid Emergency Response System for Four-Wheeler Vehicles
Abhishek N, Dr. Kavitha A S, Nandish S, Sanjana Sindol, Bhavana
Coder Buddy – An Agentic AI-Powered Python Development Assistant for Code Review, Code Generation, and Automated Mini Project Building
Sonika D, Harshitha L
MCP-Based Context-Aware System Monitoring and Threat Detection Agent
G. Priyadharshini, M.E., Balaji A, Vishnu S, Mohamed Noufal M
AI-BASED VERIFICATION OF LLM RESPONSE
Mrs. Asha Sattigeri, Lavanya R, Pavan Gowda S
Controlling Ethical Hacking: Operating System Security
Dipali Girhe, R.V. Daund
AI-Based Health Monitoring System
Ummi Habiba, Vijetha S P, C B Shekhara, Prajwal V, Anita Patil
Iot Based Patient Health Monitoring System
Basavanagowda B S, Basavaraju Bailaannavar, Gagangowda, Vijaykumar Hovale, Prof. Padmavathi N*
Emotion-Aware Adaptive Learning Systems: A Comprehensive Survey on Artificial Intelligence-Based Personalized Education
A Rashmi, A Keerthi, Harika K, Likitha K, Dr. Muhibur Rahman T.R
Automated Threat Hunting Using AI: AI-Driven Defence Strategy
Vikram G.D, K. Sharath
Medicinal Plant Identification and Classification
Anushree N N, Deepika V M
Cisix: Technology for Net-Zero Carbon Ecosystem
Darshan D Kolmi, Goutham E, Hrishikesh Bodapatti, Javeed Hussain P, Dr. Anita Patil, Vijay Kumar
Recommendation system for cloud service on Trust management services
Deepak BN, Anusha J, Krishnaveni A, Karthik R
Intelligent Crop Disease Detection Systems: A Review of Deep Learning Approaches
Siddessh K S, Anita Patil
Intelligent Fingerprint Storage and Management System for Authentication Applications on Cloud Storage
Vaasthava Sree Sai Reddy N, Subhamay Parya, Spoorti Patil, Vaishnavi Reddy, Dr.SoniaMariaD'souza
A Study on the Architecture and Operation of Intelligent Knowledge Platforms Using LLMs and AIOps
Shilpa M R, Swetha C S
AI-Based UPI Fraud Detection Using Machine Learning and Real-Time Analysis
Atul Shivaji Kamble
Open Source Software Licenses: A Comparative Analysis of GPL, MIT, and Apache Manjunath S Rakaraddi, Bhairam V Pawar, Rahul M, Raviteja Javali,
Muhibur Rahman T. R
PlacePrep - Smart DSA Learning and Placement Readiness System
Abhishek P. Nachankar, Neha P. Dhuriya, Amisha Jain, Anisha Katkade, Kumkum Laveti, Ishwari Khandwe, Amartya Choubey, Amit Gond, Atharva Dayalwar
Age Detection Using Machine Learning Techniques: A Comprehensive Review
Akshata Pravin Tatar, Nikita Balasaheb Korde, Kirti Dinkar More
AI MEDICAL CHATBOT USING MACHINE LEARNING
Tejaswini KM, Dr. Madhu H.K
Defense Framework for Runtime Monitoring and Proactive Data Safeguarding Against Emerging Ransomware Threats
Sarojini P, Reshma G, Mr. M.V. Prabhakaran M.E
Lecture Lens: AI Based Student Learning Model
Rutuja Bhute, Prajakta Koravi, Arpita Khot, Sonali Keskar, Samruddhi Jadhav, Prof. S. S. Sangewar
A Study on the Architecture and Operation of AI-Based Digital Well-Being Monitoring Systems
Rakshitha P, Swetha C S
AgroSense II: Smart Plant Disease Detection and Treatment Recommender
Hm Mujahid Pasha, B Prem Kumar, Md Mohseen, Rohit M, Dr. Anita Patil, Mr. Pavan Kumar
AI Powered Drowning Detection and Alert System for Swimming Pool
Phanindra Reddy K, Akshatha T, B Akshitha, Gouri, G Poojitha
AI-Foresight: Market Research and Trend Analyzer
Akshata Panchal, Vidya S
Department of Computer Science and Engineering, Arasu Engineering College Prevention and Detection of Botnet Attacks using Double layered machine learning Technique
S. Parvathy, S. Mounika, M. Nihidha, M. Sruthi
A Study on Multi-Location Ledger Using DPOS (DP46) in a FPL Hyundai Automobile Dealership
Kumar. Nandini, Dr. Kumarakannan.R
Animal Tracking & Detection System
Sushilkumar Dasharath Birunagi, Pranav Pandurang Bhosale, Prathmesh Sukhadev Patil, Rushikesh Satish Mulik, Prof. A.B. Jagadale
AI Powered Phishing Email Detector with Gmail Live Scanning
Mr. M. V. Prabhakaran, Mugil M U, Srikanth T
Detection of Mobile Malware (Android) using Machine Learning and Hybrid Analysis
Mrs. Uma S, Rahini R, Rudhramitra S
Insider Threat Detection System Using Machine Learning
Mr. M.V. Prabhakaran, Naveen A, Saran R, HOD, Computer Science Engineering (Cyber Security)
AI-Based First Aid Chatbot Using TF-IDF and Maximum Marginal Relevance for Efficient Symptom-Based Assistance
Vinit Sangoi, Sachi Pandya, Het Shah, Harsh Shinde, Dr. Manimala Mahato
Detection of Blackhole and Sinkhole Attacks in Wireless Sensor Networks Using a Lightweight Secure Protocol
C. Karthika and Dr. P. E. Irin Dorathy
Wallify – Make walls talk, make vibes pop.
Dr. Pravin Shinde, Saniya Shah, Vishvesh Ghadge, Moksh Gala, Akshi Gori
MCP: Multi-Tenant Chat Support System with AI Integration
Preksha Dewoolkar, Chirag Patankar, Samruddhi Pande, Dr. Rahul Pachade
VR MEDICAL TRAINING: AN IMMERSIVE AND INTERACTIVE PLATFORM FOR SURGICAL SIMULATION AND ANATOMICAL EXPLORATION
Cinita Mary Mathew, George Mathew, Alen James, Rehan Siby Joseph
CONCEPTS OF NETWORKING
Mukesh Shantaram Khodke, Mayur Goyekar, Asst. Prof. Namrata Bachhav
AI-Based Mock Interview System Using Natural Language Processing and Real-Time Feedback
Tushar Patel, Krisha Bhanushali, Drashti Gajara, Samiksha Thakur, Rahul Pachade
AQUAVISION:AI POWERED FISH SPECIES DETECTION AND DISEASE ANALYSIS
Ashrith L S, Vidya S
Smart Mirror Using Raspberry Pi
Prof. R. M. Sahu, Patil Vishal Shrikant, Parde Aditya Babasaheb, Pawar Vaishnavi
A Study on the Data Science Life Cycle and Its Applications in Modern Intelligent Systems
Samarth, Theerthashree G S
AI-Based Driver Fatigue Detection and Alert System: A Comprehensive Review
Ashith Shankar Avula, Dr. Muhibur Rahman T R*
Towards Intelligent Healthcare: A Study of AI Applications, Challenges, and Future Trends
Kaveri Suresh Boraste, Akansha Gunvantrao Deshmukh, Kirti Dinkar More
Research Design Approaches in MediNet – AI Health Risk and Smart Hospital Finder
Sainath Reddy Y S, Pani arvind, Sreenivasa Reddy, Sharath Kumar, Dr. Muhibur Rahman T.R
FitVision: Vision-Based Intelligent Fitness Assistance Using Pose-Guided Movement Analysis
Swati Uparkar, Purva Ambre, Aashna Anchan, Hasan Contractor, Husain Contractor
Retrieval-Augmented Generation for Smarter Web Scraping and Synthesis: A Cache-Aware Multimodal Framework for Web Intelligence
Rishith Poojary, Aryan Rana, Aditya Magar, Dev Raval, Pravin Shinde
Engunity X : AI-Powered SaaS Platform for Research, Code Generation and Data Analysis
Shubh Rajesh Shah, Nigranth Shailesh Shah, Shreyash Chetan Mokani, Anurag Mrityunjay Kumar, Rahul Pachade
DermAI: A Vision Transformer-Based Web System for Multi-Class Skin Disease Detection
Utilizing DINOv, Priyanka Dnyaneshwar Muley
AI-Assisted Computer-Aided Diagnosis for Early Detection of Lung Cancer
Ghanashyam Pawar, Pranav Shinde, Bharati Mahale
Copyright Issues in Digital Content Platforms
MOHAMED ZAID, A SAI SANDEEP, Dr. Muhibur Rahaman T.R
INTELLIGENT PRIVACY PRESERVING DATA ENCRYPTION AND AΝΟΝΥΜΙΖΑΤΙON SYSTEM
Sangeetha M.E., Sanjeevi R B, Vinoth M
Krishi Mitra: A Multilingual AI-Powered Conversational Agent for Indian Farmers Integrating Government Schemes, Real-Time Mandi Prices, Crop Disease Detection, and Agricultural Advisories.
Jayashri Gajre, Sujay Manasubdar, Ishan Chaudhari, Shubham Panigrahi, Suraj Patil
HEPATOSCAN:AI BASED LIVER TUMOR SEGMENTATION SYSTEM
Deekshith B N, Theerthashree G S
From Paper to Prototype: An End-to-End AI System for Automated Research Implementation
Taranjeet Singh, Monish Patil, Tejas Mungekar, Swaraj Gadre, Rupali Shinde
MedAI-DX: An AI-Powered Real-Time Clinical Decision Support System for Resource-Constrained Healthcare Environments
Hemanth Gowda A, Jayanth Somashekar, Akarsh M, Vinay Gowda PN, Dr. Kavitha AS
Early Detection of Comorbid Anxiety and Depression Using Explainable Machine Learning on DASS-21 Psychometric Data
Pranto Bosu, Satinder Kaur, Tajbir Singh
Distributed Denial of Service Attack Detection using Machine Learning
Namrata Sunil bodhale
VERIPILL:ML BASED COUNTERFEIT MEDICINE PREDICTION
Mrs. Archana N, Tarun R, Monika. K, Pranav Ramesh, Priya R K
Adaptive SPD-YOLO: Enhancing Spatial Feature Retention for Lunar Boulder Detection
Dhruv Solanki, Shashank Singh, Shrawani Sawant, Sudhanshu Singh, Ms. Swati Uparkar
A SecureStep-Smart Personal Security and Safe Navigation Platform
Dhanalakshmi S, Prajwal B, Poonamlal, Rajeev BS, Pavan Sai
AI-Based Early Detection and Risk Prediction of Jaundice Using Clinical and Liver Function Test Data
Vaseekaran A, Hariharan S, and Mrs. Malathi G
Smartbill Intelligent Retail Invoice And Profit Analysis System
P.R Ajitha M.E, MBA, Gokul P, Arvinth Kumar A
A Lightweight Wireless Intrusion Detection System for Real-Time Deauthentication and Rogue Access Point Mitigation.
Nandini Rajesh Kasar
AI-POWERED SMART INTERVIEW ASSISTANT WITH REAL-TIME FEEDBACK AND SCORE PREDICTION
Dr. Umma Habiba, Jeni Pragashini J, Kayathri A, Yogalakshmi M
AI-Powered Automated Business Report Generator
Dr. T. R. Muhibur Rahman, Haseeba Kouser, Nazia Taj S, Sai Vaishnavi D
HIDDEN IN PLAIN SIGHT: MODERN TECHNIQUES FOR PII PROTECTION
Shashank M N, Sandarsh Gowda MM
A Multi-Agent Retrieval-Augmented Generation Framework for Context-Aware Legal Document Analysis
Dr. C N Shariff, Aaftab Zohra, K Sowmya, K Rakshitha, Aishwarya G
Microstructural and Mechanical Characterization of Al2031–TiC Composites
Madevappa Channabasappa Koppal, Binit Kumar, Naresh DJ, P Ravi Kumar, Dr Santosh Janamatti, Venkatesh KC
IoT-Based Smart E - Voting System using Face Authentication
Pro. Dr. R. K. Moje, Alure Omprakash Bhagwan, Surnar Amol Nagnath, Kamble Dattatray Bharat
Comparative Study of Air-Ground Robotic Systems and Autonomous Multi-Modal Navigation Techniques
Dr Chanda V. Reddy, Maanya Arun, Khushi K, Varsha Ranganatha, Hamsa S
Real -Time Medicine Tracking and Distribution Coordination System
Ms. A. Nisha, ME., Guruprasath M, Nagoor Meeran T
Carbon Footprint Tracking in Logistics System: A Survey of AI-Based Emission Monitoring Approaches
Bhagyashri K Kulkarni, Shashank Urs H P, Vinod, Sandeep T Rathod, and Mayur Kishan Rathod
KG-CTCN: Knowledge-Guided Causal Temporal Convolutional Networks for Event-Driven Sugarcane Red Rot Forecasting
Utkarsh Joshilkar, Prasanna Bandiwadekar, Aditya Chatte, Kousen Attar, Kshitij Aiwale
Exam Section Automation for Higher Educational Institutions
Hayath T M, Yashitha C, Vidyashree, Y Indusree, Yashodha
SKY SHIELD A SKETCH-BASED DEFENSE SYSTEM AGAINST APPLICATION LAYER DDOS ATTACK
Mrs. R. Janaki, ME., Srikanth S, Krishnaraj K
Optimised Battery Thermal Management System: Design and Thermal Analysis Review
Syed Mohammd Ashfaque, C Mohhammad Asif, Shaik Naveed, Altaf
“Recipe Generation from Food Image Using Deep Learning”: A Comprehensive Review
Vijaya Durga H, Sree vishnu, Dadavali H
A Modular Data Deduplication Framework with Support for Multi-Format Document Analysis
Shreya Modi, Sreenivasa M, Shridevi, Shreya Y.P, Trisha Prameela Y
A Survey on Machine Learning and Deep Learning Techniques for Fake Review Detection
Ms. Samruddhi P. Ingale, Dr. V. H. Deshmukh, Dr. P. P. Deshmukh
AI-Powered Deepfake Detection and Liveness Detection
Dr. Jagadish R M, Chetana HK, K Shashikala, Mounika M, Pushpitha JR
Edu Site: A Advanced Learning Platform Using AI-powered Education and Personalized Learning
Kiran Mudaraddi∗, Prem Kumar B T, Prerana B, and R M Sneha
A NOVEL BRAIN-COMPUTER INTERFACE FRAMEWORK FOR REAL-TIME CONTROL AND COMMUNICATION USING NEURAL SIGNAL PROCESSING
Sara Vaivswat, M Sri Durgaa, Simran Shiva Prasad T, R K Nithish, Pranav Sankar, R Tarun Gorani, Dr Sonia Maria D’Souza
Hybrid Renewable Energy System for Medical Supply Chains
M.Lokesh, Sharanappa.Y, Kalyan Kumar.A, Karegowda.A, Jayaprakash.B
AI-BASED RESUME ANALYZER
Mrs. Sameena Yasmeen, N Indhu, Nancy Mittal, P Harika, P Saima
Semantic And Sentiment Analysis of Multilingual Code-Mixed Text
Saritha D, Bhoomika LM, Lavanya Shetty, Raksha K, Reeta Gracy
STUDY ON PROPETIES BANNANA FIBER COMPOSITIONS
Lokesh P, Manoj D P, Honnurswamy P, Srinidhi N P
“SOLAR REFRIGERATION SYSYTEM USING PELTIER MODULES”
Keshava Murthy U, Lokesh B, Manjunath B, Manjunatha K M
“Design, Development, and Analysis of a Solar Thermal Powered Automatic Air Purification System’’
Teja Raju P, Samil Raju P, Tharun Varma Raju V, Nabirashul P
Hybrid Embedding Model for Document Classification
Ranjana S. Chakrasali, Chandana A. Athreyesa, K. Shridevi B. Adiga, Vathsala, Vishnupriya
DESIGN OF AUTOMATION SEALPOT USED IN GASLINE TO REMOVE CONDESATE WATER
Mohammed Kaif, Mohammed Nawaz, Mohammed Gayaz, S Mohammed Tousif
AI-Based Smart Mobile Robotic Arm with Adaptive Gripping System
Mohammad Ibrahim, Chandan Kumar Kushwaha, Mohammed Zuhair C, Mohammad farooq
“Design & Performance Analysis of EOT Crane, Modification from DC Electromagnetic to Thruster Brake System”
Mohammed Muzameel, N Momin, Mohammed Kaif, C D Mounesh
AI-Powered Multi-Purpose Agricultural UAV for Precision Farming, Smart Crop Protection, and Real-Time Monitoring: A Comprehensive Review
Nitin Narendra Ghanmode, Rutuja B M, Rudra Prakash Tiwari, Indra Prakash Tiwari
AUTONOMOUS AGRICULTURAL ROBOT FOR SEEDING, SPRAYING, AND WEEDING
Anosh Gundi, Anand Kulagod, Hebry Sunny K, Parashuram B
A Hybrid EIGamal - AES Based Secure Image Encryption System
Gunipe Abhinav, K.V.V Subba Rao
Real-Time Hand Gesture Recognition and Sign Language Detection
Boda Deepthi, Tutta Naga Venkata Durga
Mechanical Steering-Controlled Adaptive Headlight System
Arbaz.U, Chandrashekhar.K, S. Immanviel David Raj, K. Durgashashank
REDUCTION OF WORK ROLL PEEL OFF INSTANCES IN HOT STRIP MILL
Nawal Kumar Dubey, H. Vinod Kumar, Vinay Kumar.H, Pradeepa.A
“Finite Element Analysis of Connecting Rod Stresses, Deformation, and Material Optimization”
A Kiran Kumar, A Varun Kumar, B Lingaraju, Chetan G Gotur
AI SMART CAR
Nandu G, Dr Kavitha A S, Kushal C, Hithesh Gowda, Nishanth G S
AI-Based Cyber Fraud Detection System
Mr. Harikrishna, M Shreya Reddy, Manasa V, Kirti M Maracharaddi
AUTOMATED EVALUATION OF PROGRAMMING ASSIGNMENTS
Moksud Alam Mallik, Nousheen Begum, Hafsa Yasmeen, Dr. Mujeeb Hasan
THERMOELECTRIC WASTE HEAT RECOVERY SYSTEM
Dharmana Prasad, P Sampath Kumar Achari, Penta Kunta Prudhvi, Praveen
Hybrid-Solar-Dryer
Muskan Begum, B Shambhulinga, P Pavan Kumar, V Sudheer
A Smart AI-Based Food Redistribution System for Reducing Food Waste and Supporting Community Welfare
Dr. Phanindra Reddy K, Kalyani N T, Kavya B, Keerthana P, P Sahana
Multidisciplinary Review on FEA, Topology Optimization and AI-Based Analysis of Aero-Engine Turbine Blades
Zayed Mulla, Mujaffar Hussain, Y. V. Rohinish, Mohammed Aayyon Khan
Medical Diagnosis for Coronary Arteries Disease Detection Using Deep Learning
J Aakash, M Balaji, M Syed Hussain, Dr. A Samuel Chelladhurai
Human Following Robotics
K Judah Benhur, D C Darshan Gouda, D Bhanuchandra, H Chandu.
REMOTE CONTROLLER PORTABLE DRILLING MACHINE
Pruthvi Raj RM, Rohit Sardar, Sundeep BM, Somushakar Reddy
PRODUCTION OF FUEL OUT OF PLASTIC WASTE USING PYROLYSIS PROCESS
Prajwal L, Jayashankar, G Sandeep, GV Varun
Design & Fabrication of Bio-Composites Material for Automotive / Engineering Application
Abhishek Rajbhar, Suresh B, H P J Aalibaaba, Chandra Shekar Reddy K
OPTIMIZATION OF NITROGEN TO DRY COAL BUNKERS
P. Vijay Kumar, P. Sandeepa, Tarun Kumar BC, K. Sudarshana
NAMMA BUS: Intelligent Transportation System Using Deep Learning
Dr. B. M. Vidyavathi, Suchitra M, V Rakshitha, Yamini V, Eshwar G
A Survey on Machine Learning Approach for Momentum Shift Detection and Win Prediction in Cricket Matches
Mr. J. R. Harshavardhan, Mithun M Parashar, Pavan D R, Punith Gowda G, and, Sachin N B
Secure Real-Time Coding Interview System with AI-Based Evaluation and Monitoring
Mr. J.R. Harshavardhan, Shreyas B, Yashas A, Yashwanth K, and Shiva Prakash J
Smart Civic AI: Intelligent Complaint Management System
Archana N, Yogeshwaran G, Uday D R, Manoj Kumar C, Sachin R
An Explainable AI Framework for Lifestyle-Based Healthcare Prediction: A Comprehensive Survey
Vidyasre N, Mahit Rao P, Mithun N, S G Ravidas, Shashidhara S C
“HEAT TRANSFER ANALAYSIS AND OPTIMIZATIION OF FINS BY VARIATION IN GEOMETRY”
Siddalinga K, Pinjari Asif, Murali Krishana K, C D Mounesh
Engineering Students’ Perception of AI Replacing Human Skills in Future Technical Careers: A Survey Study
Shreya.S, Tanushree. R, Srinidhi.N, Dr. Sonia Maria D'souza
AI-Powered Smart Biker Safety and Monitoring Platform
Dakshayini G R, Megha, Veena S Malipatil, Vindhya R
AI-Powered Penetration Testing Platform for Automated Vulnerability Detection: A Survey of Artificial Intelligence Methods for Identifying System Weaknesses
Mrs. Bindu K.P, Gagana R, Kushal K R, M G Sahana, and M Harshit Pramod
AI BASED INTELLIGENT MOVIE PRODUCTION AND SCRIPT GENERATOR
Dr. Samuel Chellathurai A Ph.D., Shyamvijay H, Vetrivel R
OPTIMIZING USER WORKFLOW IN AI EMAIL AUTOMATION VIA BROWSER-RESIDENT DOM MUTATION
Sabirhussen Sajidahmed Shaikh, Mahesh Chidanand Konnur, Syed Shamshud Tabrez Ahmed Shahqadri
Causal Artificial Intelligence: Modeling Cause-Effect Relationships for Intelligent Decision Systems
V. N Sheetal, Deepika V M
DOCKER: CONTAINERIZATION TECHNOLOGY
Prashanth Kumar G, Dr Madhu H K
A Literature Survey on AI-Based Secure Border Intrusion Detection Systems Using Deep Learning, IoT, and Encrypted Communication
Dakshayini G R, Manya BM, Meghana KJ, Rida Shariff, Sneha GK
Credit Card Fraud Detection Using Machine Learning
Yash Ramesh Nalawade, Harshad Sanjay Onkare, Yash Popat Bhosale
A Survey on IoT-Based Emergency Navigation and Tracking Device
Mrs. Sushmitha Suresh, Deepthi A Kumar, Inchara S, K P Nihaal, Lalith Adithya M
A Preliminary Study on a Unified Haptic Glove for Multi-Domain Applications
Mrs. Sushmitha Suresh, Ismath Zehera, Janhavi S Thorat, K G Soumya, Karanam Vennela
Design and Implementation of a Wrong Side Vehicle Detection System Using YOLO And OpenCV.
Chirag Narsinghani, Sankalp Chokale, Nikhil Teli, Pratima Chougule
AI Website Builder Using Pern Stack
Prathamesh Jalindar Yadav, Vishwajeet Milind Chavan, Mrs. S.S.Parekh
An AI-Powered Automated Code Review System Using Large Language Models and Static Analysis
Mrs. Nita Meshram, Ganni Naveen Raj Anudeep, K Vedavyas, K Harsha Nandhan, C Balaji Naidu
Optimized Random Forest Model for Chronic Kidney Disease Classification with Imbalanced Data Handling
Kapil, Ankit Navgeet Joshi
AI-Powered Smart Platform for Government Scheme Eligibility Using RPA
Archana N, Vaibhavi S, Tejaswini P, S Akshatha, Rishmitha K B
SafeDox: A Review on Secure Document Sharing Using Blockchain, IPFS, and End-to-End Encryption
Neeharika Shakthivelan, Badareesh P, C S Jeevan Setty, Poorvi P, Mrs. Poornima H N
Model Context Protocol (MCP): A Standard Interface for Tool-Aware AI Systems
Prajwal P, Prof. Swetha C S
CTR Prediction and Campaign Simulation System using Machine Learning
Litrishiya Merceline Mary A, Nivetha S, Dr. K. Ravikumar
AI-SMART LEARN PERSONALISED LEARNING
Dr. K. Prem Kumar, K.Ashritha, J.Amruth, M.Usha Rani, B.Siddhartaa, S.Karunakar Reddy
Explainable Multimodal AI for Deepfake Detection and Digital Content Authenticity
R S Geethanjali, Sneha KS, Vaidehi Vasudev Arundekar, Vinutha BR
AI-Based Emergency Response System Using Facial Recognition, GPS Navigation and Number Plate Recognition
C J Nagasri Pragna, C Panduranga Reddy, Chaitra Patil, D Eshwar Kumar, Dr. Renuka Sagar
Hybrid Quantum-Classical Optimization for Energy Distribution using TSP Model
Mrs. Punitha M R, Aditya H, Ankitha P, Anusha M N, and Deekshitha K
AI-BASED EARLY POULTRY DISEASE RISK PREDICTION SYSTEM USING REAL-TIME TREND ANALYSIS
Mamatha R, Bhavya Sai Shree V, Chithra U, Deeksha N, Dyuthi S
“A Survey on Prompt Injection and Jailbreak Defenses in Large Language Models”
Anushree K N, Vedanth M, Amulya H, Chirayu Gowda, Prof. Meghashree C
AI-Powered Detection of Deepfake Audio in Hindi and Kannada Using Speech Analysis
Mrs. Kavitha K S, Chaitanya C Gowda, D Yashwanth, Dheeraj R, Lishanth N
Innovations in Stroke Identification: A Machine Learning-Based Diagnostic Model Using Neuroimages
Malti Punjaram Barve, Prof.Puspendu Biswas
Augmented Reality (AR) Based Virtual Lab: A Survey on Interactive AR Learning Platforms and Educational Applications Using Deep Learning Techniques
Meena G, Mayurakhi Maiti, Sheethal M, Shilpa B, and Vineela K
AI Model to Detect Bone Fracture
Kavitha K S, Mallikarjun Biradar, Pavan S, Sourabh Goud Allolli, Yallaling Meti
EMOTRIXA: AI Companion Robots for Emotional Support
Rishika. B, Dr Kavitha AS
Violence Detection in Video Surveillance Using Frame Extraction
Priyanka Sunil Kognole, Shruti Mahadev Pisal, Suparshwa Sudhir Kognole
Smart Flood Detection and Impact Mapping System: An IoT-Driven Framework for Real-Time Early Warning and Spatial Risk Visualisation
Mrs. Vidya V Patil, Amar, Divit V, Hari Narayana S, Kiran S
AI-BASED SMART FARMING MONITORING SYSTEM
Anushya V P, D Nikhil, Darshan M, Dodagatta Mahesh, Nitish Kumar
Automated Blood Cell Segmentation and Classification Using YOLOv11n: An End-to- End Deep Learning Approach
Dr. Jagadish R M, Poorvi V Mallapur, Pragna Kakandaki, Rakshitha Atti
Smart Anti-Theft SmartPhone Ecosystem With Offline Remote Access Using BLE and GSM: A Survey of Hardware and Algorithmic Methods for Securing Devices and Automated Offline Recovery
Mrs. Bindu K.P, Dimple J, Gabburi Narasanna Pallavi, Gaddamamadugu Dinavya
Smart Crop Doctor: An AI Driven Chilli Plant Disease Detection
Dr. C N Shariff, A Pravalika, D Ranjitha, D Vindhu, Ganga
AI Powered Smart Banking Assistant
Dr. K Venkata Rao, Arpita Patil, Chaithanya R, Harshitha S, Jyothi P
Privacy-Preserving Intelligent Collaboration Platform with Searchable Encryption and Client Side AI: A Survey of Privacy-Preserving Methods for Secure Collaboration
Dr. Kothapalli Venkata Rao, Priya H M, Saniya Fatima, and Varsha R
Smart Parcel Receiving System
Jayashubha J, Ashwini N R, Amrutha K, A Yashwitha, K Bindu
Wildlife Poaching Risk Prediction and Detection Using Satellite AI
Mrs. Meena G, Impana P, Keerthana P, Harshitha M
Nature-Inspired Deep Learning Approach for Cardiac Arrest Detection
S. M. Bokey, S. R. Gupta, R. R. Karwa
A Comparative Benchmark of Deep Learning Models and Deployment of a Web Application for Automated Early Heart Attack Risk Prediction
Ms. R.Devi, Dr. K. Padma Priya
Countify: A Text-to-Image Generation Model
Prof. Dr.S.S.More, Athrav Raju Ugale, Vaidehi KeshavVaze, Vaishnav Bajrang Patil, Alisha Mubarak Attar, Sanika Appasaheb Patil
Real-Time Physiological Monitoring and Automated Alert System for Enhanced Soldier Safety and Survivability
Manjunath Kammar, Anushya V P, Dr. Jagadish RM, Ganta Poojith Balu, S Mounika
Micro-Grid Energy Management Using Inverters
Manoj Gowda B P, B J Jayadeva, Prajwal H R, Vaibhav S
AI-Based Smart Bus Live Tracking System Using IoT and Machine Learning: A Survey of Real-Time ETA Prediction and Transportation Analytics Approaches
Jayashubha J.G, Muppuri Dhatri, Raavi Pranitha, Rayapati Supriya
AI-Based Intelligent Document a Verification System Using OCR and Machine Learning
Vrushali Gavali, Rutuja Shelar, Pranita Sandim
Artificial Intelligence Techniques in Clinical Decision Support Systems in Radiology Using X-ray and MRI
Shubhmehar Singh Dhillon*, Sandeep Kaur, Satveer Kour
AI Powered Mental Health Companion
Hasini P, Jestina M, D.Sasmitha M.E
AI-Based Virtual Interview Assistant: An Intelligent NLP-Driven System for Automated Interview Evaluation and Performance Analysis
Anitha, Ankith Kumar Verma, Archana Kulkarni, B S Shree Roopa, Dr. Phanindra Reddy K
A Comparative Analysis of Lightweight Deep Learning Models for Real-Time Apple Orchard Monitoring
Mansi Sharma
Green Elixir Vision: An Intelligent Ayurvedic Herbal Recommendation Platform for Personalized Natural Healthcare
H. Srinivas, J. Vinay, K. Basavana Gouda, K. G Ravi Teja Gowda, Dr. C. N. Shariff
Web Security Issues Analysis and Practical Ways to Prevent Them
Praveen kumar, Manjeet kaur
A Research on Machine Learning Driven Gamification Model for Personalized Education
Sadiya Ali, Dr. G. R. Bamnote, Dr. G. J. Sawale
QR CODE BASED CAFETERIA FOOD ORDERING AND PAYMENT SYSTEM
Dr. V. U. Bansude, Pathan Surayya, Mane Gauri, Patale Mayuri
A Machine Learning-based Crop Yield Forecasting and Recommendation System
C M Abhishek, Channaveeresha Meti, Revanasiddappa, Mallikarjun Kallappa Poojari, Dr. Jagadish R M, Manjunath Kammar
A Review Study of Quantum Computing
Gurpreet Kaur, Satveer Kour, Sandeep Kaur
From Intuition to Agriculture 4.0: A Comprehensive Review of Internet of Things (IoT) and Machine Learning in Precision Farming
Maryen, Satinder Kaur
Beyond Standalone Classifiers: A Critical Review of Multi-Paradigm and Ensemble Machine Learning Architectures in Crop Recommendation Systems
Maryen, Satinder Kaur
ROLE AND CHALLENGES OF HARITHA KARMA SENA: A STUDY IN ERNAKULAM DISTRICT
Dr. Francis M C, Anosh Ignatious, Diya Damian and Dr. Somasekharan T M
INTELLIGENT SCREAM DETECTION SYSTEM FOR CRIME ALERTS
Dr.Hemavati C Purad, Putti Monashree, Pawan Kumar N, Sai Ganesh M
AI‑ML Powered Robot for OCR‑Driven Payment Reconciliation using UiPath
Ashutosh Mankar, Dr G.R Bamnote, Prof. S.P Akarte
Design And Implementation of An Embedded CNN based Weed Detection and Mechanical Weed Elimination Rover
Sai Badgujar, Raj Patil, Gaurav Shinde, Parth Borgude, Rupali Shirsath, Prof. Mayur Kumbharde
AI-Based Sign Language Recognition System
Dr. K. Prem Kumar, S. Pranavi, T. Manojghna, P. Manaswini, Y. Bhavitha
Predictive Analysis of Health and Lifestyle Patterns using CRM and Machine Learning
Prathamesh Chambole, Dr S.R Gupta, Dr R.A Kale
A Lightweight HSV Histogram-Based Algorithm for Real-Time Face Recognition on Edge Devices
Prof. Roshni Gawande, Dr. S. B. Patil, Prof. Sneha Dhere
Kanwariya Sewa: A Smart Logistics and Wellness Platform for Pilgrimage Support Using Real-Time Tracking and Emergency Response System
Subhash, Swati Sinha, Shrey Agnihotri, Dr. Deepak Kumar Gupta
Comparative Performance Analysis of Quantum Error Correction Codes under Simulated Depolarizing Noise using Qiskit
Gurpreet Kaur, Satveer Kour, Sandeep Kaur, Satinder Kaur
Zero Trust Architecture in Cloud Security: Principles, Implementation, and Challenges
Varun Kumar, Sandarsh Gowda M M
Social Engineering Attacks in Cybersecurity: Analysis, Challenges, and AI-Based Defense Framework
Vikas Gowda J V, Prof. Swetha C S
A Comparative Analysis of Machine Learning Algorithms for the Early Prediction of Heart Disease
Manan Mehra
Social Engineering Attacks in the AI Era: A Review
Dr. Sangeeta Rani, Mr. Gopal Sharma, Dr. Kapil Kumar Kaswan
Database Reliability and Availability Across Multi-Cloud Environments
Naresh Kumar Miryala
Identification of Leukemia Subtypes from Microscopic Images Using ResNeXt Algorithm
Mr. Arul Leo Felix L, Ms. Shereen J, Ms. Sushmitha S, Mr. Vishva S
Data-Driven Billing Reconciliation and Workforce Analytics via UiPath and Power BI
Ashutosh Mankar, Dr G.R Bamnote, Prof. S.P Akarte
Brain Tumor Detection and Classification using Deep Learning
Vaibhavi Ghorpade, Sanika Patekar, Sanika Satre and Gauri Wakchaure, Prof. Dr. Sachine Bere, Mr. A. M. Suryawanshi
Evaluating Symmetric Encryption: Performance and Security Considerations for Enterprise Data Protection
Naresh Kumar Miryala, Meta Platforms Inc. CA, USA
SPECTROSCOPIC BIOMARKER DETECTION FOR URINE DISEASES USING MACHINE LEARNING
Mr. Surendhiran S Mr. Saran K, Ms. Savina S, Mrs. Nirupashri G
Energy Harvesting and Health Monitoring Device
Dr.D.O.Patil, Dr.R.K.Moje, Sanika Jagnade, Aditi Bhosale, Trupti Sonavane
AI-DRIVEN INTELLIGENT POWER GRID OPERATION FOR A SMART AND SUSTAINABLE ENERGY FUTURE
Spandana H A, Thrisha S, Thanuja M U, Spoorti Irappa Chamakeri, Dr. Sonia Maria D'souza, Prof.Manojkumar, Ms.Apeksha N H
Improving MNIST Image Synthesis via an Optimized Generative Adversarial Network with Transfer Learning and Real-Time Loss Monitoring
Dr.C.Deepa, Dr.B.Vidhya, Dr.V.Sumathi, Dr.N.Mahendiran, Dr.M.Praneesh
AI-POWERED PERSONAL HEALTH RECOMMENDATION SYSTEM
Abdoulie Bawo, Sajin Tamang, Modou Lewis, Aakash Ranjan, Sahil Kumar, Dr. P. M. Gavali
AI- Powered Public Safety Monitoring System
Anand Tasgave,Sahil Thomake,Aditya Vhanuagre,Shahid M.Ali Vijapure,Harshwardhan Patil, Prof. K. S. Kadam
Micro Front End Architecture: Accelerating Team Scalability in Modern Tech
Ritika B Immadi, A G Vishvanath
AI-Driven Space Debris Detection, Tracking and Removal System
Mrs. Nita Meshram, M Pooja, Prarthana J, Priyanka Kumari, Spandana S P
Speech- Driven note-taking with AI-Based Transcription, Translation and Summarization
Khushi h Dhongadi, Swetha M
An AI-Powered Integrated Placement Module with Automated Proctoring and Chatbot-driven Interview Simulation for Higher Education
Sayli Patil*, Chaitali Banmare, Pranoti Nule, Diksha Thakar, Pratiksha Titkare
OncoSight – AI Powered Early Cancer Detection System
Dr. Samuel Chellathurai A.Ph.D, Sibimathavan K, Yogaraj M
Brain Box: An AI-Powered Multimodal Knowledge Organizer Platform
Ashitosh Sanjay Langare, Snehal Namdev Lohar, Sanika Sunil Mane, Rohit Dilip Patil, Prof. Mansi Khanaj
BCI Performance Optimization Using EEG Band Power and Hybrid Deep Learning
Dr.H.Umma Habiba, Sarani M, Arthi J, Shamli S
GEOSTONE DETECTION AI – INTELLIGENT ROCK AND GEMSTONE RECOGNITION SYSTEM
Dr. Samuel Chellathurai A, Vaishnavi S, Subalakshmi R, Thamarai R
AMR INSIGHT: MACHINE LEARNING-BASE MICROBE RESISTANCE PREDICTION USING ARFA
Dr. Swetha Singh, Subash M, Akash V
AI-Enabled Sensor-Based Crop Stress Early Prediction and Detection System
Kanishkumar R, Sivasankar D, Dr. E. Sivanantham
INTELLIGENT ENERGY SAVING SYSTEM FOR HOME APPLIANCES
Mrs. Nigileeshwari, Sivaguru A, Mohamed Irfan A, Dhinesh S
AN INTELLIGENT CULINARY INSTRUCTION GENERATOR FOR AUTOMATED RECIPE CREATION FROM VISUAL FOOD IMAGES
K.S Sukrutha, Shruthi H.R, Anusha C.G, Sowbhagya.P, Varsha.S, Varshini.R-
AI BASED CROP MARKET PRICE PREDICTION SYSTEM USING MACHINE LEARNING
Mr. Muthukrishnan M.E, Vishwa S, Navin Kumar K
Predicting Cleft Lip in Unborn Babies Using Ultrasound Images and Machine Learning
Mrs. Dakshayini G R, Mahalakshmi P S, Dheekshith M, Devish Papani, Prajwal koushik C
Application of Mohan Transform in Initial and Boundary Value Problems
Bhatendra kumar*, Jyotsna chandel
Shuffling in Artificial Intelligence: Foundations, Methodologies, Applications, and Future Directions
Ahmed S. AlMahmeed
An Artificial Intelligence-Driven Framework for Text Similarity Measurement and Plagiarism Detection Using Hybrid Lexical and Semantic Analysis
DIDDE PRAVEEN KUMAR, A.N. RAMA MANI*
An AI-Driven Framework for Real-Time Energy Consumption Monitoring, Demand Forecasting, and Optimization Using Deep Sequence Models
M. Lavanya Durga, K. Lakshamana Reddy*
Serverless Document Workflow Orchestration, Cloud-Native Architecture Using AWS Step Functions for Distributed Approval Processing
Veeravalli Jyothika, K. Lakshmi Sai Sri*
A Scalable, Serverless Cloud Architecture for Real-Time Courier and Shipment Tracking Using AWS ECS Fargate and Containerized Microservices
BOKKA DIVYA, Dr. CHIRAPARAPU SRINIVASA RAO*
A Privacy-Preserving AI Companion for Student Mental Wellness: Adaptive Conversational Support with Sentiment-Aware Monitoring and Retrieval-Augmented Guidance
KONALA NAGA SOWMYA SREE, Mr.B.N. SRINIVASA GUPTA*
A Machine Learning Framework for Procrastination Detection and Academic Performance Prediction with Local Large-Language-Model Feedback
SANJANA ROKKALA, P SRINIVASA REDDY*
Abstract
Data-Driven Analysis of Mortality and Survival Outcomes in Patients with Liver Cirrhosis: A Global Healthcare Analytics Study
Mehdi Mostofi, Prasanna Poreddy, Jayasri Katragadda, Nadeem Basha Syed, Sunandamala Bodaballa
DOI: 10.17148/IJARCCE.2026.15501
Abstract: Liver cirrhosis is a progressive chronic disease associated with high morbidity and mortality, requiring accurate risk stratification to support clinical decision-making. This study presents a data-driven analytical framework integrating survival analysis, machine learning, and uncertainty quantification to improve mortality prediction in cirrhosis patients. A retrospective dataset of 418 patients was analyzed using Kaplan–Meier estimation and Cox proportional hazards modeling to evaluate survival patterns and identify significant predictors. For predictive modeling, multiple supervised learning algorithms—including Logistic Regression, Decision Tree, Random Forest, Support Vector Machine, Gradient Boosting, and Extreme Gradient Boosting (XGBoost)—were implemented to classify patients into high- and low-risk groups. Model performance was assessed using accuracy, recall, F1-score, and ROC-AUC, with cross-validation employed to ensure robustness. To enhance reliability, conformal prediction was applied to quantify predictive uncertainty at a predefined confidence level. Results indicate that disease stage, age, bilirubin, and albumin are significant predictors of mortality. Ensemble models demonstrated superior predictive performance, with XGBoost achieving the highest recall and strong discrimination. Conformal prediction provided well-calibrated uncertainty estimates, improving the interpretability and trustworthiness of model outputs. The findings demonstrate that integrating statistical and machine learning approaches enhances mortality risk stratification and supports the development of reliable clinical decision-support systems.
Keywords: Liver Cirrhosis; Survival Analysis; Machine Learning; Mortality Prediction; Cox Proportional Hazards; XGBoost; Risk Stratification; Conformal Prediction
Keywords: Liver Cirrhosis; Survival Analysis; Machine Learning; Mortality Prediction; Cox Proportional Hazards; XGBoost; Risk Stratification; Conformal Prediction
Abstract
An Effective Heart Disease Detection and Severity Level Classification Model Using Machine Learning and Hyperparameter Optimization Methods
Dr. Ganesh.G. Taware, Miss.P.D. Nale, Kaveri Kalbhor, Gaurav Karande, Swaraj Navale, Yogesh Gaikwad
DOI: 10.17148/IJARCCE.2026.15502
Abstract: The rapid advancement of Machine Learning (ML) techniques has enabled the development of intelligent systems for early disease prediction. This research presents a web-based Heart Disease Prediction System that leverages ML algorithms to estimate the risk of cardiovascular disease using patient health parameters. The system is developed using the Django framework and integrates a trained Random Forest Classifier to analyze clinical inputs uch as age, blood pressure, cholesterol level, and other relevant medical attributes.
The proposed model processes user-provided data and generates a probability score, which is further categorized into Low, Borderline, and High-risk levels. The system also incorporates rule-based adjustments to handle ex- treme cases, thereby improving the reliability of predictions. Additionally, all predictions are stored in a data- base, enabling users to track their health assessment his- tory over time.
The experimental evaluation demonstrates that the model achieves satisfactory accuracy and provides quick and ac- cessible results through a user-friendly interface. Although the system is not intended to replace professional medical diagnosis, it serves as an effective preliminary screening tool for raising awareness and encouraging early medical consultation. This work highlights the potential of integrating machine learning with web technologies to build scalable and accessible healthcare support systems.
The proposed model processes user-provided data and generates a probability score, which is further categorized into Low, Borderline, and High-risk levels. The system also incorporates rule-based adjustments to handle ex- treme cases, thereby improving the reliability of predictions. Additionally, all predictions are stored in a data- base, enabling users to track their health assessment his- tory over time.
The experimental evaluation demonstrates that the model achieves satisfactory accuracy and provides quick and ac- cessible results through a user-friendly interface. Although the system is not intended to replace professional medical diagnosis, it serves as an effective preliminary screening tool for raising awareness and encouraging early medical consultation. This work highlights the potential of integrating machine learning with web technologies to build scalable and accessible healthcare support systems.
Abstract
Smart Hybrid Bus Tracking System with Mobile Sensing and Real-Time Updates
Anil Kumar T R, Harshavardhan Pattar, Indusri J, Joshi Chetan, Dr. Yeresime Suresh, Dr. Anita Patil
DOI: 10.17148/IJARCCE.2026.15503
Abstract: Growing urban populations have created an increasing need for intelligent public transportation systems, with real-time bus tracking becoming a critical requirement. Traditional approaches rely extensively on specialized GPS equipment, creating significant costs and operational challenges. This research presents a Smart Hybrid Bus Tracking System that utilizes smartphones that drivers and passengers already carry, removing the requirement for dedicated hardware. The system uses the driver's mobile device as the main location provider, with passenger smartphone data enhancing position precision. During network outages, the platform saves information locally and updates it when connectivity returns. The system also integrates sophisticated algorithms such as Kalman Filter, XGBoost, and LSTM to improve prediction accuracy. This approach delivers a cost-effective, flexible, and practical tracking solution that works effectively across various real-world applications.
Keywords: Bus Tracking, GPS, Mobile Sensing, Real-Time System, Kalman Filter, XGBoost, LSTM, Smart Transportation
Keywords: Bus Tracking, GPS, Mobile Sensing, Real-Time System, Kalman Filter, XGBoost, LSTM, Smart Transportation
Abstract
Smart Health Tracker with Doctor Consultation
Satyam Sahu, Jitendra Gupta, Suraj Verma, Mr. Kumar Bibhuti
DOI: 10.17148/IJARCCE.2026.15504
Abstract: Healthcare is undergoing a quiet but consequential shift. For generations, patients have relied on physical visits to hospitals and clinics for diagnosis, consultation, and record management—a model that works well for some individuals and poorly for many others. The rise of digital technologies and intelligent systems offers a practical way out of this limitation. This paper presents a Smart Health Tracker with Doctor Consultation platform that collects patient health data, analyses it efficiently, and enables continuous monitoring, record management, and remote consultation. Rather than treating health tracking and doctor consultation as separate services layered on top of conventional systems, we integrate them into a unified architecture from the ground up. The platform comprises multiple tightly coupled modules: a data collection layer, a health record management system, a report analysis component, a doctor consultation module, and an administrative dashboard. A functional evaluation with users demonstrated improved accessibility to healthcare services, better management of medical records, and more efficient interaction between patients and doctors compared with traditional healthcare approaches. The system is modular and cloud-deployable, making it easy to extend or integrate with existing healthcare infrastructure.
Keywords: Smart Health Tracker; Telemedicine; Artificial Intelligence; Health Monitoring; Doctor Consultation; Medical Report Analysis; Digital Healthcare; Web Application
Keywords: Smart Health Tracker; Telemedicine; Artificial Intelligence; Health Monitoring; Doctor Consultation; Medical Report Analysis; Digital Healthcare; Web Application
Abstract
Smart Food Donation System: A Web-Based Approach for Efficient Food Redistribution–A Review
Akshay, Prashanth K, K.Shiva darshan, Kishan kumar B.V, Pavan kumar P, Anitha Patil P
DOI: 10.17148/IJARCCE.2026.15505
Abstract: Food waste continues to be a serious global issue even as many communities face hunger and food insecurity. This review presents a Smart Food Donation Management System (SFDMS) - a web-based platform that connects food donors with recipients efficiently. Donors like restaurants, households, and institutions can post details of available surplus food, while receivers such as NGOs and shelters can request and collect it easily. The system supports location- based matching, tracks expiry dates, and provides built-in communication features for better coordination. An administrative dashboard enables monitoring, report generation, and data management. Testing results demonstrate the system to be user-friendly, reliable, and effective in reducing food waste while improving food distribution.
Keywords: Food Donation, Web Application, Waste Reduction, NGO Support.
Keywords: Food Donation, Web Application, Waste Reduction, NGO Support.
Abstract
AI-Based Video surveillance System
Afsa Saboo, B Sai Dikshitha, C Mohammad Athiq, K Sudeep Gouda, Nagateja P, Anita Patil
DOI: 10.17148/IJARCCE.2026.15506
Abstract: Rapid urbanisation and escalating public safety demands have created an urgent need for intelligent, automated surveillance solutions that can operate without continuous human oversight. Conventional CCTV infrastructure places excessive cognitive load on operators monitoring multiple feeds simultaneously, increasing the risk of missed incidents due to fatigue and delayed response. This paper proposes a dual-model AI surveillance framework that concurrently detects three real-world emergency categories — road accidents, fire incidents, and suspicious human activity — by combining YOLOv8 spatial object detection with ResNet-50 temporal activity classification in a unified processing pipeline. On emergency detection, the system autonomously assembles an alert payload containing a timestamped snapshot, GPS-tagged camera location, and event confidence score, dispatching notifications in parallel via SMS, email, and mobile push notification to relevant authorities. Experimental evaluation on publicly available benchmark datasets yields detection accuracies of 89.2%, 91.5%, and 86.0% for accidents, fire, and suspicious activity respectively, with per- frame inference latency of 0.8–1.2 seconds and end-to-end alert delivery within three seconds. The proposed framework significantly reduces reliance on manual monitoring and offers a scalable, deployable foundation for smart city infrastructure, transportation hubs, and public safety control rooms.
Keywords: Artificial intelligence; video surveillance; YOLO; CNN; emergency detection; computer vision; real-time alerts; smart cities; deep learning.
Keywords: Artificial intelligence; video surveillance; YOLO; CNN; emergency detection; computer vision; real-time alerts; smart cities; deep learning.
Abstract
Smart AI-Integrated Exam Security Gate
A Sunitha, B Lavanya, B Pallavi, Manisha Patel, Anita Patil
DOI: 10.17148/IJARCCE.2026.15507
Abstract: Examination systems face major challenges such as malpractice, unauthorized entry, and the use of prohibited electronic devices. Traditional manual checking methods are often time-consuming, require more manpower, and may fail to detect hidden devices or impersonation attempts effectively. With the advancement of technologies such as Artificial Intelligence (AI), Machine Learning (ML), and sensor-based systems, smarter security solutions can be developed for examination environments. This project presents a Smart AI-Integrated Exam Security Gate that combines object detection and meta-detection sensors to provide automated verification and security screening at exam hall entry points. The system focuses on key features, such as student identity verification, prohibited item detection, alert generation, and real-time monitoring. It aims to improve the efficiency, accuracy, and reliability of the examination process while reducing human effort and security risks. This study also highlights the limitations of existing manual security systems and emphasizes the need for a more intelligent, automated, and secure examination management solution.
Keyword: Artificial Intelligence (AI), Exam Security, Object Detection, Metal Detection Sensor, Student Verification, Real-Time Monitoring, Smart Security Gate, Alert System.
Keyword: Artificial Intelligence (AI), Exam Security, Object Detection, Metal Detection Sensor, Student Verification, Real-Time Monitoring, Smart Security Gate, Alert System.
Abstract
SMART RAILWAY TRACK HEALTH MONOTORING SYSTEM by IOT
Dr. C.N. Deshmukh, Mayur S. Salode, Aryan R. Kale, Minal G. Lonkar, Kshitij M. Thotange
DOI: 10.17148/IJARCCE.2026.15508
Abstract: Undetected railway track anomalies, such as surface cracks and tilting, pose severe safety risks to modern transportation networks. To address this vulnerability, this paper presents a Smart Railway Track Health Monitoring System utilizing Industrial Internet of Things (IIoT) principles for continuous, real-time structural assessment. Powered by an ESP32 microcontroller and the ESP RainMaker cloud platform, the framework autonomously scans for physical faults, alignment shifts, and leveling irregularities. Sensor telemetry is processed locally and routed directly to the cloud, pushing instant, percentage-based severity alerts to remote control room personnel. By replacing reactive manual inspections with an automated predictive pipeline, this low-cost system significantly accelerates maintenance response times, minimizes operational downtime, and ensures a safer railway infrastructure.
Keywords: IoT, ESP32, Railway Crack Detection, Tilt Sensor, VibrationSensor, Arduino Uno, Track Monitoring, Railway Safety
Keywords: IoT, ESP32, Railway Crack Detection, Tilt Sensor, VibrationSensor, Arduino Uno, Track Monitoring, Railway Safety
Abstract
Farm2Door: A Smart Digital Platform for Farmer-to-Customer Agricultural Marketplace-A Review
K Naveen, K Shyam, Kothapalli Vamshi, Kothapalli Vardhan Babu, Dr. Anita Patil, Asst.Prof. Shashikantha Raddi
DOI: 10.17148/IJARCCE.2026.15509
Abstract: The demand for fresh agricultural products has increased significantly in recent years, but the existing supply chain still faces multiple inefficiencies. In traditional systems, farmers depend on intermediaries to sell their produce, which often leads to reduced profits and higher prices for customers. To address this issue, this paper presents Farm2Door, a digital platform that directly connects farmers with consumers.
The platform allows farmers to upload their products and manage availability, while customers can browse, compare, and place orders easily. Features such as order tracking, simple payment options, and customer feedback are included to improve usability and trust. By reducing the dependency on middlemen, the system helps in maintaining fair pricing and better transparency.
Although the model is simple, it has the potential to improve local agricultural trade if implemented effectively. Future improvements can include smarter delivery systems and better demand prediction.
I
The platform allows farmers to upload their products and manage availability, while customers can browse, compare, and place orders easily. Features such as order tracking, simple payment options, and customer feedback are included to improve usability and trust. By reducing the dependency on middlemen, the system helps in maintaining fair pricing and better transparency.
Although the model is simple, it has the potential to improve local agricultural trade if implemented effectively. Future improvements can include smarter delivery systems and better demand prediction.
I
Abstract
CLEARBON: A Blockchain-Based Dual Market System for Real-Time Carbon Credit Verification and Trading
Sohail, Sooraj B, Suma Shree S, Swetha Rani H S, Anita Patil
DOI: 10.17148/IJARCCE.2026.15510
Abstract: Industrial carbon markets face delays in credit issuance and lack real-time verification and predictive trading mechanisms. This paper proposes CLEARBON, a blockchain-based system integrating IoT sensors, AI models, and satellite data for real-time emission monitoring and instant carbon credit issuance. The system introduces a dual-market approach with spot and futures trading, enabling efficient, transparent, and secure carbon transactions while supporting predictive emission reduction strategies.
Keywords: Blockchain, Carbon Credits, IoT, Machine Learning, Smart Contracts, Emission Monitoring
Keywords: Blockchain, Carbon Credits, IoT, Machine Learning, Smart Contracts, Emission Monitoring
Abstract
EVOLUTION OF AI TECHNOLOGIES IN THE EDUCATIONAL SECTOR
Yashvardhan Chahal, Yash Goswami, Tushar Mahaur, Yash Jain, Dr. K.S. Mishra
DOI: 10.17148/IJARCCE.2026.15511
Abstract: Artificial Intelligence (AI) has rapidly evolved from early rule-based systems to today’s advanced machine learning models, significantly transforming the global educational landscape. Over the years, AI technologies such as adaptive learning platforms, intelligent tutoring systems, virtual assistants, predictive analytics, and automated assessments have revolutionized how students learn and how teachers manage academic tasks. This research paper explores the historical evolution, technological progression, applications, challenges, and future potential of AI-driven educational solutions. Despite issues such as data security, cost, and ethical concerns, AI continues to reshape education, making learning more personalized, inclusive, and efficient. The study concludes that AI technologies will play a central role in building next-generation smart learning environments.
Keywords: Artificial Intelligence, Educational Technology, Intelligent Tutoring Systems, Personalized Learning, Machine Learning.
Keywords: Artificial Intelligence, Educational Technology, Intelligent Tutoring Systems, Personalized Learning, Machine Learning.
Abstract
Talent Track Workforce Employment Solutions Platform
Sathiyaa Balajee R, Mrs. Saranya R
DOI: 10.17148/IJARCCE.2026.15512
Abstract: The Talent Track Workforce Employment Solutions Platform is a comprehensive digital system designed to streamline and modernize the recruitment and workforce management process. It serves as an integrated platform that connects job seekers, employers, and administrators, enabling efficient communication, skill matching, and hiring decisions. The system aims to reduce the complexity of traditional recruitment methods by providing a centralized environment where users can manage job postings, applications, and candidate profiles with ease. It also supports data organization and reporting, enabling better decision-making and workforce planning. The platform is designed with scalability and flexibility in mind, making it suitable for organizations of varying sizes and industries.
Keywords: Employer Dashboard, Resume Management, Skill Matching, Online Recruitment, Hiring Process Automation, Web-Based Application, Job Seekers, Employers, Talent Acquisition, Data Management, Performance Tracking, Scalable System, User-Friendly Interface.
Keywords: Employer Dashboard, Resume Management, Skill Matching, Online Recruitment, Hiring Process Automation, Web-Based Application, Job Seekers, Employers, Talent Acquisition, Data Management, Performance Tracking, Scalable System, User-Friendly Interface.
Abstract
SMART EXAM SEAT ALLOCATION SYSTEM USING ANDROID APPLICATION
Riya Bhoir, Bhumika Beldar, Kanishka Chindarkar, Prof. Smita Chunamari
DOI: 10.17148/IJARCCE.2026.15513
Abstract: Smart Exam Seat Allocation System is an Android-based application designed to automate the process of assigning seats to students during examinations. The system reduces manual effort, ensures fairness, and optimizes classroom utilization. It integrates modules such as login authentication, classroom creation, student data management, seat generation, and student view. This paper discusses the motivation, objectives, system architecture, methodology, technology stack, features, advantages, limitations, and future scope of the system.
Keywords: Seat Allocation, Android Application, Java, Firebase, Automation, Classroom Management, Exam System
Keywords: Seat Allocation, Android Application, Java, Firebase, Automation, Classroom Management, Exam System
Abstract
“Smart AI-Based Student Attendance System with Monthly Analytics”
Harsh Sharma, Miss. Taniya Jain, Dr. Uruj Jaleel, Dr. Satish Kumar Soni
DOI: 10.17148/IJARCCE.2026.15514
Abstract: Attendance management is a critical administrative function in educational institutions, playing a vital role in monitoring student participation, discipline, and academic performance. Accurate attendance records are essential for evaluating student engagement and ensuring compliance with institutional policies. However, traditional attendance systems, including manual registers and biometric-based systems, suffer from several limitations such as time inefficiency, susceptibility to human errors, proxy attendance, and lack of real-time monitoring and analytics. With the rapid advancement of emerging technologies such as Artificial Intelligence (AI), Machine Learning (ML), and Computer Vision, there is a growing demand for intelligent systems capable of automating repetitive tasks while providing meaningful insights. In this context, facial recognition technology has emerged as a powerful tool for identity verification and automation.
This research proposes a Next-Generation Smart AI-Based Student Attendance System integrated with Monthly Analytics and Predictive Insights. The system utilizes real-time facial recognition techniques to automatically identify students and mark attendance without requiring manual input or physical interaction. The integration of deep learning algorithms ensures high accuracy and robustness under varying environmental conditions. In addition to attendance automation, the proposed system incorporates a comprehensive analytics module that processes attendance data to generate monthly reports, identify trends, and predict future attendance behavior. These predictive insights enable educators and administrators to identify students at risk due to low attendance and take proactive measures. The system is implemented using Python, OpenCV, and advanced machine learning techniques, ensuring scalability and efficiency. Experimental results demonstrate that the proposed system significantly reduces time consumption, eliminates proxy attendance, and enhances data reliability. This research highlights the potential of combining AI- driven automation with data analytics to transform traditional attendance systems into intelligent decision-support systems, contributing to the development of smart education environments.
Keywords: Artificial Intelligence, Face Recognition, Computer Vision, Attendance Analytics, Machine Learning, Smart Education.
This research proposes a Next-Generation Smart AI-Based Student Attendance System integrated with Monthly Analytics and Predictive Insights. The system utilizes real-time facial recognition techniques to automatically identify students and mark attendance without requiring manual input or physical interaction. The integration of deep learning algorithms ensures high accuracy and robustness under varying environmental conditions. In addition to attendance automation, the proposed system incorporates a comprehensive analytics module that processes attendance data to generate monthly reports, identify trends, and predict future attendance behavior. These predictive insights enable educators and administrators to identify students at risk due to low attendance and take proactive measures. The system is implemented using Python, OpenCV, and advanced machine learning techniques, ensuring scalability and efficiency. Experimental results demonstrate that the proposed system significantly reduces time consumption, eliminates proxy attendance, and enhances data reliability. This research highlights the potential of combining AI- driven automation with data analytics to transform traditional attendance systems into intelligent decision-support systems, contributing to the development of smart education environments.
Keywords: Artificial Intelligence, Face Recognition, Computer Vision, Attendance Analytics, Machine Learning, Smart Education.
Abstract
AI-Driven Personalized Education Platform: Design, Architecture, and Implementation
Sajid Raza, Vishal Kumar Sah, Rishabh Tiwari, Saif Siddiqui, Dr. Sandeep Kumar Dubey
DOI: 10.17148/IJARCCE.2026.15515
Abstract: Education is undergoing a quiet but consequential shift. For generations, classrooms have delivered the same lesson to every student at the same pace—a model that works well for some learners and poorly for many others. The rise of machine learning offers a practical way out of this impasse. This paper presents an AI-Driven Personalized Education Platform that collects fine-grained learner data, analyses it in real time, and continuously adapts the content, assessments, and feedback each student receives. Rather than treating personalisation as a premium feature layered on top of a conventional system, we built it into the architecture from the ground up. The platform comprises five tightly coupled modules: a data collection layer, an AI analysis engine, a recommendation module, an adaptive assessment component, and a teacher-facing dashboard. A semester-long pilot with undergraduate Computer Science students showed improved quiz performance, faster identification of weak topics, and more targeted teacher interventions compared with a static content pathway. The codebase is modular and cloud-deployable, making it straightforward to extend or integrate with existing institutional infrastructure.
Keywords: Personalized Learning; Artificial Intelligence; Machine Learning; Adaptive Assessment; Recommendation System; Educational Data Mining; E-Learning.
Keywords: Personalized Learning; Artificial Intelligence; Machine Learning; Adaptive Assessment; Recommendation System; Educational Data Mining; E-Learning.
Abstract
Artificial Intelligence in Cybersecurity: A Comprehensive Survey on AI-Driven Insider Threat Detection
Deepak Kumar G., Deevaraj M., H Pramodh, Lakshmi Narayana, Dr. Muhibur Rahman T R
DOI: 10.17148/IJARCCE.2026.15516
Abstract: Insider threats are one of the most difficult cybersecurity problems organizations face today. Unlike attacks that come from outside, insider threats involve people who already have authorized access — employees, contractors, or trusted partners — who either deliberately misuse that access or unknowingly create security risks. Because these individuals operate within normal system boundaries, traditional security tools like firewalls tend to miss them entirely. This paper looks at how artificial intelligence is being used to tackle this problem, drawing on published research from IEEE Xplore, ACM Digital Library, Springer, and ScienceDirect. We looked at a range of approaches — deep learning, graph-based analysis, User Behavior Analytics (UBA), Support Vector Machines, rule-based methods, and even psychosocial behavioral modeling. To make sense of this variety, we put together a four-tier framework that organizes these systems from the simplest rule-based tools all the way up to fully adaptive AI platforms. We also measured how these systems perform in terms of detection accuracy, false alarm rates, scalability, and speed. One finding kept coming up: no existing system brings together real-time monitoring, automated risk scoring, explainable outputs, and adaptive learning in a single working platform. We explore why this gap exists and what it would take to close it.
Keywords: Artificial Intelligence; Cybersecurity; Insider Threat Detection; Machine Learning; Deep Learning; User Behavior Analytics; Anomaly Detection; Graph-Based Detection; Support Vector Machine; LSTM; Risk Scoring; Behavioral Analysis; Explainable AI; Real-Time Monitoring; Intrusion Detection Systems.
Keywords: Artificial Intelligence; Cybersecurity; Insider Threat Detection; Machine Learning; Deep Learning; User Behavior Analytics; Anomaly Detection; Graph-Based Detection; Support Vector Machine; LSTM; Risk Scoring; Behavioral Analysis; Explainable AI; Real-Time Monitoring; Intrusion Detection Systems.
Abstract
A Comprehensive Study on Data Storage Security Issues and Services in Cloud Computing
Sujal Satish Godse, Bhushan Sunil Matsagar, Kirti Dinkar More
DOI: 10.17148/IJARCCE.2026.15517
Abstract: Cloud computing has become an important platform for storing and managing data due to its scalability, flexibility, and cost effectiveness. However, outsourcing data to third-party cloud service providers introduces several security concerns, particularly related to confidentiality, integrity, and availability. This paper presents a study of data storage security issues in cloud computing and discusses cloud service models and deployment models from a storage perspective. It reviews existing security techniques such as encryption, identity-based authentication, and third-party auditing for ensuring data protection. The paper also examines major challenges including data privacy, data recoverability, media sanitization, insecure APIs, vendor lock-in, and network dependency. The study highlights the need for effective security mechanisms to improve trust and reliability in cloud storage environments.
Keywords: Cloud computing, Cloud storage, Data security, Zero Trust, Blockchain, Confidential computing
Keywords: Cloud computing, Cloud storage, Data security, Zero Trust, Blockchain, Confidential computing
Abstract
SMART ROAD ACCIDENT DETECTION AND RESCUE SYSTEM
JEEVITHA R, B SAHANA REDDY, G TEJASHWINI, SHREYA B R
DOI: 10.17148/IJARCCE.2026.15518
Abstract: Road accidents are a major global issue, leading to significant loss of life due to delayed emergency response. This paper proposes a Smart Accident Detection and Rescue System that uses IoT sensors, GPS, and communication technologies to automatically detect accidents and alert emergency services. The system integrates accelerometers, gyroscopes, GPS modules, and GSM communication to identify collisions and transmit real-time location data. Advanced approaches using machine learning further improve detection accuracy and reduce false alarms. The proposed system ensures faster rescue operations, minimizes response time, and enhances road safety.
Abstract
MultiCart- Ai Based Multi Vendor Cart Platform
Kamlesh Kumar Pal, Abhishek Gupta, Harshit Mall, Mr. Deepak Kumar
DOI: 10.17148/IJARCCE.2026.15519
Abstract: The rapid growth of e-commerce has created a need for scalable and efficient platforms that can support multiple vendors within a single ecosystem. Traditional online shopping systems are often limited to single vendors, restricting product variety and reducing operational flexibility. To overcome these limitations, this project presents “MultiCart- Ai Based Multi Vendor Cart Platform,” an intelligent web-based application designed to integrate multiple sellers and provide a seamless shopping experience for users.
The proposed system introduces a unified cart mechanism that allows customers to add and purchase products from different vendors in a single transaction. It incorporates Artificial Intelligence (AI) techniques to enhance user experience through personalized product recommendations, smart filtering, and behavior-based suggestions. The platform also provides dedicated dashboards for vendors to manage produc t s, t r ack orde rs, and analyze performance, while administrators can monitor system activities, approve vendors, and maintain platform integrity.
The system is developed using modern web technologies such as React.js for the frontend, Node.js and Express for the backend, and MongoDB for database management. Secure authentication and efficient data handling ensure reliability and scalability of the platform.
The implementation of Multicart demonstrates i m prove d usa bi l i t y, e f fi c i ent vendor m a na ge me nt, and enhance d custome r satisfaction compared to traditional systems. This project highlights the potential of combining multi-vendor architecture with AI-driven features to build a next-generation e-commerce platform.
The architecture of Multicart follows a modular and layered design, promoting flexibility, maintainability, and efficient data handling. The implementation results demonstrate that the system effectively manages multi-vendor operations, reduces redundancy, and provides a smooth and intelligent shopping experience.
Furthermore, the platform addresses key issues in existing systems, such as lack of personalization, inefficient cart management, and limited scalability.
The proposed AI-powered multi-vendor cart platform offers a robust and future-ready solution for modern e-commerce applications. It not only improves operational efficiency for vendors and administrators but also enhances the overall user experience through intelligent automation and seamless integration of services. Future enhancements may include advanced machine learning models, mobile application support, and secure payment gateway integration.
The proposed system introduces a unified cart mechanism that allows customers to add and purchase products from different vendors in a single transaction. It incorporates Artificial Intelligence (AI) techniques to enhance user experience through personalized product recommendations, smart filtering, and behavior-based suggestions. The platform also provides dedicated dashboards for vendors to manage produc t s, t r ack orde rs, and analyze performance, while administrators can monitor system activities, approve vendors, and maintain platform integrity.
The system is developed using modern web technologies such as React.js for the frontend, Node.js and Express for the backend, and MongoDB for database management. Secure authentication and efficient data handling ensure reliability and scalability of the platform.
The implementation of Multicart demonstrates i m prove d usa bi l i t y, e f fi c i ent vendor m a na ge me nt, and enhance d custome r satisfaction compared to traditional systems. This project highlights the potential of combining multi-vendor architecture with AI-driven features to build a next-generation e-commerce platform.
The architecture of Multicart follows a modular and layered design, promoting flexibility, maintainability, and efficient data handling. The implementation results demonstrate that the system effectively manages multi-vendor operations, reduces redundancy, and provides a smooth and intelligent shopping experience.
Furthermore, the platform addresses key issues in existing systems, such as lack of personalization, inefficient cart management, and limited scalability.
The proposed AI-powered multi-vendor cart platform offers a robust and future-ready solution for modern e-commerce applications. It not only improves operational efficiency for vendors and administrators but also enhances the overall user experience through intelligent automation and seamless integration of services. Future enhancements may include advanced machine learning models, mobile application support, and secure payment gateway integration.
Abstract
Public Transport Tracking Systems: A Comprehensive Study of Real-Time Solutions for Small Cities
C Renuka, Divya Rajeev M, J Kavyashree, Bindu AV, Muhibur Rahman T R
DOI: 10.17148/IJARCCE.2026.15520
Abstract: Public Transport Tracking Systems leverage advanced location-based technologies to enable accurate real- time vehicle monitoring and efficient transit management, departing from traditional systems that rely on fixed schedules and limited passenger information. As urban mobility evolves—particularly in small cities with growing population and infrastructure constraints—the need for scalable, cost-effective, and accessible real-time transport solutions has intensified. This paper presents a structured review of multiple studies from IEEE, Springer, ScienceDirect, and related sources, covering core technologies such as GPS-based tracking, IoT-enabled transport systems, mobile application integration, and cloud-based data processing. A novel four-tier taxonomy is proposed, classifying systems based on functional capabilities: real-time vehicle tracking, passenger information systems, fleet management and optimization, and smart transport assistant platforms. Performance aspects including tracking accuracy, latency, system reliability, scalability, and cost-efficiency are analyzed. Comparative evaluation reveals that no existing solution fully integrates real-time tracking, predictive arrival estimation, route optimization, and user-interactive platforms into a unified system suitable for small cities. Several research gaps are identified, and a strategic roadmap toward intelligent, affordable, and integrated public transportation systems is outlined.
Abstract
CLICKET – Be There or Make It Happen With AI
Sharad Singh, Ritik Chaurasiya, Shivam, Prince Yadav, Kumar Bibhuti Bhushan Singh
DOI: 10.17148/IJARCCE.2026.15521
Abstract: The rapid advancement of digital technologies has significantly transformed the way events are organized, managed, and experienced across various domains such as education, business, entertainment, and social engagement. Most existing systems focus on isolated functionalities such as ticket booking or event listing, without providing a comprehensive solution that integrates all stages of the event lifecycle. As a result, users often face challenges related to event discovery, complex booking processes, and lack of centralized management, while organizers struggle with operational inefficiencies and the need to rely on multiple tools.
A key innovation of the proposed system lies in its integration of artificial intelligence to enhance user experience and system efficiency. The AI-powered category suggestion feature analyzes event descriptions and automatically recommends appropriate categories, improving event discoverability and reducing the effort required by organizers. Additionally, the system incorporates intelligent workflows that enable better organization and classification of events, thereby improving search relevance and user engagement.
Keywords: Event Management System, Artificial Intelligence, QR Code Verification, Next.js, MongoDB, RBAC, Web Application
A key innovation of the proposed system lies in its integration of artificial intelligence to enhance user experience and system efficiency. The AI-powered category suggestion feature analyzes event descriptions and automatically recommends appropriate categories, improving event discoverability and reducing the effort required by organizers. Additionally, the system incorporates intelligent workflows that enable better organization and classification of events, thereby improving search relevance and user engagement.
Keywords: Event Management System, Artificial Intelligence, QR Code Verification, Next.js, MongoDB, RBAC, Web Application
Abstract
AWAZE-E-JANATA- VOICE OF THE PEOPLE
Rajveer Pratap Singh, Aman Tiwari, Raj Yadav, Mr. Dileep Kumar Gupta
DOI: 10.17148/IJARCCE.2026.15522
Abstract: Awaze-e-Janata – Voice of the People is a digital platform designed to bridge the communication gap between citizens and government authorities. The primary objective of this system is to provide a simple, accessible, and transparent medium through which people can raise their complaints, issues, and suggestions related to public services. This platform enables users to submit complaints manually or with the assistance of Artificial Intelligence, making the process faster and more efficient. The system categorizes and forwards the complaints to the relevant authorities, ensuring that the issues reach the correct department without delay. Additionally, it allows users to track the status of their complaints, promoting transparency and accountability. The project focuses on solving real-world societal problems by empowering citizens and enhancing public participation in governance. By integrating modern technologies such as web development and AI-based assistance, the system aims to improve communication, reduce response time, and build trust between the public and government institutions.
Overall, Awaze-e-Janata serves as an effective tool for digital governance, ensuring that every voice is heard and every issue is addressed efficiently.
Overall, Awaze-e-Janata serves as an effective tool for digital governance, ensuring that every voice is heard and every issue is addressed efficiently.
Abstract
Smart Battlefield Helmet with Soldier Monitoring System
Dr. Pramod Sharma, Aryansh Gupta, Divya Gupta, Jayanth Bharadwaj, Sagar Mishra
DOI: 10.17148/IJARCCE.2026.15523
Abstract: This paper presents a Smart Battlefield Helmet with Soldier Monitoring System. The aim of the project is to build a low-cost, embedded system that can continuously monitor a soldier’s health and surroundings and transmit the data wirelessly to a base station. The helmet uses an Arduino Nano as the main microcontroller, paired with an ESP32 for communication. Sensors include the MAX30102 for heart rate and SpO₂ monitoring, DS18B20 for body temperature, an MQ gas sensor for detecting toxic or flammable gases, and an MPU6050 IMU for fall detection. A NEO-6M GPS module provides real-time location tracking, and a LoRa SX1278 radio module transmits all data to a base station dashboard several kilometres away. The circuit runs on a 5 V regulated supply using an LM7805 regulator powered from a 9 V DC adapter.
Keywords: Helmet, Soldier, Monitoring, Temperature, Tracking, Battlefield
Keywords: Helmet, Soldier, Monitoring, Temperature, Tracking, Battlefield
Abstract
AUTOFACE - Attendance Simplified Through Vision
Shrestha Gupta, Shreoshi Roy
DOI: 10.17148/IJARCCE.2026.15524
Abatract: AutoFace: Attendance Simplified
through Vision is an automated attendance management system designed to overcome the
limitations of traditional manual methods, such as inefficiency, human error,
and proxy attendance. The proposed system leverages deep learning–based facial
recognition using the SSD MobileNet v1 architecture for real-time face
detection under varying conditions. Detected faces are encoded into
128-dimensional embeddings and matched against a secure database using the
Euclidean Distance metric for accurate identity verification.
Developed using the MERN stack, the system ensures
scalability, real-time data synchronization, and platform independence. It also
provides features such as live session monitoring and automated report
generation. Experimental results demonstrate an accuracy of 97.8% with an
average latency of less than 1.5 seconds per individual. The system offers a
secure, contactless, and efficient solution, significantly improving
reliability and reducing administrative overhead in attendance management.
Keywords: AutoFace,
Face Recognition, Attendance Management System, Deep Learning, SSD MobileNet,
Facial Embeddings, Euclidean Distance, MERN Stack, Cloud-Based System,
Real-Time Monitoring
Abstract
DejaView: An AI-Powered Academic Project Integrity and Evaluation Portal Using Vector Embeddings and Large Language Models
Mrs. Trupthi Rao, Mr. Sumit Kumar Yadav, Aditya Raj Swain, Mahadev Jagtap, Pratik, Gajanan
DOI: 10.17148/IJARCCE.2026.15525
Abstract: The proliferation of student project submissions in higher education institutions has introduced acute challenges in maintaining academic integrity and evaluation consistency. Manual review processes are labour-intensive, subjective, and fail to scale with the growing volume of submissions. This paper presents DejaView, a full-stack, AI-powered academic project management portal developed for the Department of Computer Science and Engineering (AI & ML) at Dayananda Sagar University. DejaView automates two critical pain-points: (1) detection of plagiarised or over-similar submissions using 384- dimensional dense vector cosine similarity powered by the FastEmbed BAAI/bge-small-en-v1.5 model, and (2) structured generation of executive summaries, technology stack labels, and personalised viva voce questions using the Groq LLM (llama- 3.3-70b-versatile). The system is built on a modular three-layer Python architecture comprising a Streamlit frontend, an AI sentinel engine, and a thread-safe SQLite persistence layer. Institutional access is enforced through strict email domain gating and bcrypt password hashing, ensuring that only verified university stakeholders interact with the platform. Experimental results demonstrate that the system reliably flags submissions with greater than 75% cosine similarity, generates domain-relevant AI insights with high contextual precision, and exports structured grade sheets for archival in a single click.
Keywords: Academic Integrity, Plagiarism Detection, Vector Embeddings, Large Language Models, Federated Evaluation, Streamlit, Cosine Similarity, Viva Question Generation, FastEmbed, Groq API.
Keywords: Academic Integrity, Plagiarism Detection, Vector Embeddings, Large Language Models, Federated Evaluation, Streamlit, Cosine Similarity, Viva Question Generation, FastEmbed, Groq API.
Abstract
EMPLOYMENT HIRING PLATFORM USING AI
Sanjay Kumar, Prince Pandey, Md Shahjad, Deepak Kumar
DOI: 10.17148/IJARCCE.2026.15526
Abstract: This paper presents JobVista, a full stack employment hiring platform designed to support students, job seekers, and recruiters through a role-aware digital recruitment workflow. The platform provides secure authentication, candidate profile management, recruiter company management, job posting, job browsing, application tracking, applicant review, and assistant-driven guidance through JobMate. The implemented system uses React, Vite, Redux Toolkit, Express.js, Node.js, MongoDB, Mongoose, JWT-based cookie authentication, and bcrypt password hashing. JobVista also improves early platform usefulness by showing selected external opportunities and by providing template and Gemini-backed assistant responses for resumes, interviews, cover letters, job posts, and screening questions. The paper discusses the problem background, proposed methodology, architecture, database design, module implementation, testing, results, limitations, and future enhancements. The final system demonstrates that a college- level job portal can be technically meaningful, visually usable, and extensible without becoming unnecessarily complex.
Keywords: JobVista, job portal, employment hiring platform, React, Node.js, Express.js, MongoDB, role-based access control, application tracking, JobMate, AI career assistant
Keywords: JobVista, job portal, employment hiring platform, React, Node.js, Express.js, MongoDB, role-based access control, application tracking, JobMate, AI career assistant
Abstract
A Review of Voice-Driven Accessible Vocational Training Platforms for Persons with Visual Impairment
Bharath Reddy, Chinta Vishnu Vardhan, H M Sharanabasava, K Ravi Kumar, Dr. Muhibur Rahman
DOI: 10.17148/IJARCCE.2026.15527
Abstract: The increasing demand for inclusive education and skill development has brought significant attention to assistive technologies designed for persons with visual impairments. Voice-driven vocational training platforms represent a transformative approach to delivering accessible learning experiences without reliance on visual interfaces. This review examines existing literature, system architectures, and technological frameworks used in voice-based e-learning and vocational training tools tailored for visually impaired users. The paper presents a four-tier taxonomy of platform sophistication, analyses key methods including Text-to-Speech (TTS), Automatic Speech Recognition (ASR), Natural Language Processing (NLP), and screen reader integration, and identifies critical research gaps. Comparative analysis of reviewed systems highlights strengths and limitations. The review concludes by recommending a unified, adaptive voice- driven platform with offline capabilities and multi-language support to bridge the vocational accessibility gap.
Keywords: Visual Impairment, Voice-Driven Learning, Vocational Training, Assistive Technology, Text-to-Speech, Speech Recognition, Accessible Education, NLP
Keywords: Visual Impairment, Voice-Driven Learning, Vocational Training, Assistive Technology, Text-to-Speech, Speech Recognition, Accessible Education, NLP
Abstract
NeuroSecure: A Comprehensive Survey on Deep Learning Approaches for Cyber Defense and Intrusion Detection
P. Anirudh, B. Adarsh Reddy, G. Raghavendra, K. Naveen Kumar, Dr. Muhibur Rahman T R
DOI: 10.17148/IJARCCE.2026.15528
Abstract: Cyber threats have grown dramatically in both scale and sophistication, outpacing the detection capabilities of classical signature-based and rule-driven security tools. This paper surveys the evolution of Intrusion Detection Systems (IDS) from their early reliance on static pattern matching through to modern deep learning–driven architectures, drawing on peer-reviewed publications and benchmark evaluation studies. We review work spanning core algorithmic approaches—Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Transformer architectures, and ensemble methods—alongside their application to network traffic analysis using the NSL-KDD and CICIDS2017 benchmark datasets. To bring structure to this body of literature, we introduce a four-tier classification framework organized around increasing system sophistication: from basic signature matching through anomaly detection and hybrid approaches, to fully integrated, context-aware deep learning platforms. We then present a hybrid ensemble IDS framework that combines CNN, RNN, and Transformer models through a majority-voting fusion layer, achieving detection accuracy exceeding 96% with precision, recall, and F1-score consistently above 94% across all attack categories. Performance dimensions examined include classification accuracy, precision-recall balance, generalization to unseen threats, and scalability for enterprise and cloud environments. A recurring observation across reviewed studies is the absence of any single system that simultaneously handles diverse attack types, class imbalance, feature redundancy, and real-time traffic volumes within one coherent architecture. We discuss the practical implications of this gap and outline directions for future research.
Keywords: Intrusion Detection Systems; Deep Learning; Convolutional Neural Networks; Recurrent Neural Networks; Transformer Models; Network Security; NSL-KDD; CICIDS2017; Ensemble Learning; Cybersecurity; Anomaly Detection; Zero-Day Attacks.
Keywords: Intrusion Detection Systems; Deep Learning; Convolutional Neural Networks; Recurrent Neural Networks; Transformer Models; Network Security; NSL-KDD; CICIDS2017; Ensemble Learning; Cybersecurity; Anomaly Detection; Zero-Day Attacks.
Abstract
Case Study on Experimental Setup Design in Robotics Research: A Comprehensive Review of Future Directions
M A Kaif, Md Akif Bari, Md Muntaseeb, Punith Raj, Dr. Muhibur Rahaman T.R
DOI: 10.17148/IJARCCE.2026.15529
Abstract: Robotics research has become a rapidly evolving domain with applications spanning industrial automation, healthcare, autonomous systems, and smart environments. A critical component of robotics research is the design of experimental setups, which ensures reliable testing, reproducibility, and performance evaluation of robotic systems. This paper presents a comprehensive case study on experimental setup design in robotics research, covering methodologies such as simulation-based testing, hardware prototyping, real-time validation, and hybrid experimental approaches. The study explores key application areas including autonomous navigation, robotic manipulation, human-robot interaction, and swarm robotics. It also highlights major challenges such as hardware constraints, environmental variability, sensor inaccuracies, cost limitations, and reproducibility issues. Furthermore, the paper identifies research gaps and proposes future directions involving AI-integrated robotics, digital twins, edge computing, and adaptive experimental frameworks. The findings emphasize that a well-designed experimental setup significantly enhances the reliability, scalability, and real-world applicability of robotics systems.
Keywords: Robotics Research, Experimental Design, Autonomous Systems, Simulation, Hardware Prototyping, Sensors, Control Systems, Human-Robot Interaction, AI Robotics, Edge Computing, Swarm Robotics
Keywords: Robotics Research, Experimental Design, Autonomous Systems, Simulation, Hardware Prototyping, Sensors, Control Systems, Human-Robot Interaction, AI Robotics, Edge Computing, Swarm Robotics
Abstract
A Review on Machine Learning Techniques for Crop Yield Prediction
A Renukamma, Arathi C G, Aruna B, K Ananya, Dr. Muhibur Rahman T.R
DOI: 10.17148/IJARCCE.2026.15530
Abstract: Crop yield prediction plays a crucial role in ensuring food security, efficient resource management, and sustainable agricultural planning. With the rapid advancement of artificial intelligence, machine learning (ML) techniques have emerged as powerful tools for predicting crop productivity using diverse datasets such as weather conditions, soil characteristics, and remote sensing data. This paper presents a comprehensive review of machine learning techniques applied to crop yield prediction. It analyzes commonly used algorithms, including linear regression, decision trees, random forests, support vector machines, and deep learning models such as artificial neural networks and convolutional neural networks. Studies show that environmental factors like temperature, rainfall, and soil type are the most significant features influencing prediction accuracy.
Keywords: Crop Yield Prediction, Machine Learning, Precision Agriculture, Remote Sensing, Predictive Modeling
Keywords: Crop Yield Prediction, Machine Learning, Precision Agriculture, Remote Sensing, Predictive Modeling
Abstract
AI-Based Real-Time Traffic Congestion Prediction and Signal Optimization System
B Deepika, Basavarajeshwari, D R Pallavi, D Suhasini, Dr. Muhibur Rahman T.R
DOI: 10.17148/IJARCCE.2026.15531
Abstract: Traffic congestion has become a major challenge in urban areas due to rapid population growth and increasing vehicle density. Conventional traffic signal systems operate on fixed timing schedules and fail to adapt to real-time traffic conditions, leading to inefficient traffic flow and increased delays. This paper proposes an AI-based real-time traffic congestion prediction and adaptive signal optimization system to address these limitations. The system utilizes techniques from Machine Learning to analyze traffic parameters such as vehicle density, time of day, and historical traffic patterns. A predictive model is developed to classify congestion levels, and based on the predicted output, signal timings are dynamically adjusted to optimize traffic flow. The proposed approach improves traffic efficiency by reducing vehicle waiting time and minimizing congestion at intersections. Experimental results demonstrate that the AI-based system outperforms traditional fixed-time signal control methods in terms of accuracy and overall traffic management performance. This work contributes to the development of intelligent and scalable solutions aligned with modern Smart City initiatives.
Keywords: Artificial Intelligence, Traffic Congestion Prediction, Adaptive Signal Control, Real-Time Traffic Management, Machine Learning, Smart Cities, Intelligent Transportation Systems, Data Analytics
Keywords: Artificial Intelligence, Traffic Congestion Prediction, Adaptive Signal Control, Real-Time Traffic Management, Machine Learning, Smart Cities, Intelligent Transportation Systems, Data Analytics
Abstract
Crop Disease Detection Using AI: A Comprehensive Survey
G Sai Supriya, G Sanjana, Harshita S S, V Indu, Muhibur Rahman T.R
DOI: 10.17148/IJARCCE.2026.15532
Abstract: Crop disease detection is critical for ensuring food security and improving agricultural productivity, especially in India where a large population depends on farming. Traditional disease identification relies on manual inspection by farmers, which is often inaccurate, time-consuming, and leads to delayed treatment and excessive pesticide use. In recent years, Artificial Intelligence (AI) and deep learning, particularly Convolutional Neural Networks (CNN), have emerged as effective solutions for automated plant disease diagnosis.
This project presents an AI-based crop disease detection system that uses CNN for identifying diseases in tomato and potato leaves from uploaded images. The system provides instant classification of Healthy or Diseased leaves along with treatment recommendations and preventive measures. A web-based interface is developed using Flask for easy access by farmers with basic digital skills. The system also maintains a history of past detections to help users track disease patterns over time.
The study evaluates the system’s performance in terms of accuracy, usability, cost-effectiveness, and scalability. Results show that the system delivers fast and reliable predictions, reduces dependency on agricultural experts, and minimizes crop loss. The project highlights the potential of AI in making agriculture more efficient and sustainable, while also identifying scope for future enhancements like multi-crop support and mobile integration.
Keywords: Crop Disease Detection; Artificial Intelligence; Convolutional Neural Network; CNN; VGG-16; Flask; TensorFlow; PlantVillage Dataset; Smart Agriculture; Image Processing.
This project presents an AI-based crop disease detection system that uses CNN for identifying diseases in tomato and potato leaves from uploaded images. The system provides instant classification of Healthy or Diseased leaves along with treatment recommendations and preventive measures. A web-based interface is developed using Flask for easy access by farmers with basic digital skills. The system also maintains a history of past detections to help users track disease patterns over time.
The study evaluates the system’s performance in terms of accuracy, usability, cost-effectiveness, and scalability. Results show that the system delivers fast and reliable predictions, reduces dependency on agricultural experts, and minimizes crop loss. The project highlights the potential of AI in making agriculture more efficient and sustainable, while also identifying scope for future enhancements like multi-crop support and mobile integration.
Keywords: Crop Disease Detection; Artificial Intelligence; Convolutional Neural Network; CNN; VGG-16; Flask; TensorFlow; PlantVillage Dataset; Smart Agriculture; Image Processing.
Abstract
Development Of Eye Disease Detection Using Deep Learning
Afrin Amin Makandar, Pranav Sanjay Kumbhar
DOI: 10.17148/IJARCCE.2026.15533
Abstract: Catching eye diseases early stops patients from going blind. Right now, doctors check eye photos by hand. This manual process takes too much time and causes frequent mistakes. This paper proposes a robust computer- aided diagnosis (CAD) system designed to automatically classify retinal fundus images into eight distinct categories: Normal, Glaucoma, Diabetes, Cataracts, Age-related Macular Degeneration (AMD), Hypertension, Pathological Myopia, and other abnormalities. Utilizing a Multi-class Fundus Image Dataset, the research implements a deep learning framework centered on Convolutional Neural Networks (CNNs). The methodology integrates advanced image preprocessing—including grayscale conversion, noise filtration, and contrast-limited adaptive histogram equalization (CLAHE)—to enhance diagnostic features. The system is deployed via a high-performance web interface, ensuring low computational overhead and seamless user interaction. Experimental results indicate high precision and recall, demonstrating the system’s efficacy in facilitating rapid, large-scale ocular screenings.
Keywords: Deep Learning, Convolutional Neural Networks (CNN), Retinal Fundus Imaging, Computer-Aided Diagnosis (CAD), Medical Image Processing, Ocular Pathology
Keywords: Deep Learning, Convolutional Neural Networks (CNN), Retinal Fundus Imaging, Computer-Aided Diagnosis (CAD), Medical Image Processing, Ocular Pathology
Abstract
AUTONOMOUS/TELE-OPERATED LANDMINE DETECTION ROBOT
Deepak Yadav, Kundan Gautam, Sarfaraz Ansari, Shailesh Srivastava, Roshni dubey, Prof. Prabhakar dubey
DOI: 10.17148/IJARCCE.2026.15534
Abstract. An autonomous/teleoperated land mine detection robot vehicle is an advanced robotic system designed to enhance safety and efficiency in detecting buried landmines in hazardous environments. The primary objective of this system is to minimize human involvement in minefield exploration, thereby reducing the risk of injury or loss of life. The robot is equipped with a combination of sensors such as a metal detector for identifying metallic components of landmines and an ultrasonic sensor for obstacle detection and navigation. It operates in two modes: autonomous mode, where the robot follows a pre-programmed path and makes decisions independently, and teleoperated mode, where it is controlled remotely by an operator using wireless communication technologies like Bluetooth or RF modules. The system is typically built around a microcontroller such as Arduino, which processes sensor data and controls actuators like DC motors through a motor driver. Additional modules such as GPS and GSM can be integrated to provide real-time location tracking and alert notifications when a potential mine is detected. Upon detection, the robot stops, activates visual and audio alerts, and transmits the location information to the operator. This dual-mode operation ensures flexibility in different terrains and operational conditions. Overall, the land mine detection robot offers a cost-effective, reliable, and safer alternative to manual detection methods, with significant applications in military operations, border security, and post-conflict area clearance, while also providing a foundation for future advancements using artificial intelligence and improved sensing technologies.
Abstract
AGRITRADEHUB: A MERN Stack Based Digital Platform for Advancing India's Farmer-Merchant Direct Trade Ecosystem
Mukesh Gupta, Dheeraj Singh, Nidhi Shukla, Roshni Soni, Ms. Shreoshi Roy
DOI: 10.17148/IJARCCE.2026.15535
Abstract: Digital platforms are rapidly reshaping industries globally, and agriculture is no exception. In India, where agriculture plays a pivotal role in the economy and livelihoods of millions, the adoption of web technologies for direct farmer-merchant trading holds immense potential to revolutionize the agri-trade ecosystem. This paper proposes Agritradehub, a MERN stack based platform that eliminates intermediaries and facilitates direct market linkages between farmers and merchants. The study explores how MongoDB, Express.js, React.js, and Node.js can be used to build a scalable, secure, and user-friendly marketplace. From real-time product listing and bidding systems to integrated payment gateways and data analytics for price discovery, the paper analyzes the transformative impact of this platform on market access, supply chain efficiency, and overall transparency. It also highlights how the system differs from government initiatives like eNAM by providing complete price control to farmers. Furthermore, the paper addresses the challenges associated with digital adoption in the agricultural sector, including the digital divide, smartphone penetration, and the need for multilingual interfaces. This paper aims to provide valuable insights for students, researchers, and practitioners involved in promoting digital transformation in Indian agriculture.
Keywords: Agriculture, MERN Stack, Digital Marketplace, Direct Trading, Farmer-Merchant Platform, E-Agriculture, MongoDB, React.js.
Keywords: Agriculture, MERN Stack, Digital Marketplace, Direct Trading, Farmer-Merchant Platform, E-Agriculture, MongoDB, React.js.
Abstract
CONTENT FILTER BASED MOVIE RECOMMENDATION SYSTEM USING AI AND ML
Asharf Khan, Alok Kumar Singh, Vishal Singh, Akrar Khan, Pankaj Kumar Gupta
DOI: 10.17148/IJARCCE.2026.15536
Abstract: Recommendation system refers to the system which, dependent on the particular data set, provides users with recommendations for certain resources, such as books, movies, songs and so forth. Generally speaking, movie recommendation system usually suggests films which the User would love to watch due to particular attributes of the User, as well as previously accumulated data. Recommendation systems are very beneficial for companies which collect their data from a large number of customers and need to provide them with the most appropriate recommendations possible. Several aspects can be considered when building a movie recommendation system, such as the movie genre, the actors, its plot and even the director. Recommendation systems can offer movies based on a single aspect or several combined ones. The recommended system used in our paper is constructed based on the tags generated from the combinations of the film genre, actors and description which may be interesting for the User to watch. We decided to apply Content-based Filtering approach in our work.
Keywords: Movie Recommendation Systems, Content-Based Filtering, Movie recommendation, ai and machine learning project.
Keywords: Movie Recommendation Systems, Content-Based Filtering, Movie recommendation, ai and machine learning project.
Abstract
Alzheimer’s Disease Detection Using AI
Divyanshi Kashyap, Divya pandey, Mansi Pandey, Divyanshi Yadav, Dr. Nikhat Akhtar
DOI: 10.17148/IJARCCE.2026.15537
Abstract: Alzheimer’s disease is a progressive neurological condition that impacts memory, thinking ability, and speech functions. Early detection of Alzheimer’s is very crucial because timely intervention can help slow the progression of symptoms and enhance the overall quality of life for patients. Conventional diagnostic methods mainly depend on neuroimaging, cognitive assessments, and clinical evaluations, which are often costly, time-consuming, and not easily accessible to all individuals.
Recent studies indicate that speech patterns and language usage can provide early indications of cognitive decline. Individuals affected by Alzheimer’s disease frequently show variations in speech fluency, vocabulary diversity, pause pat-terns, and sentence formation. These variations can be exam-ined using computational methods along with machine learning techniques.
In this work, we present a machine learning-based approach for identifying Alzheimer’s disease through speech recordings and linguistic analysis. Acoustic features such as Mel Frequency Cepstral Coefficients (MFCC), pitch, energy, speech rate, and pause duration are derived from audio signals. Linguistic features are obtained from transcribed speech using TF-IDF representation.
Multiple models, including text-based, audio-based, and combined feature models, are tested using the XGBoost classifier. The experimental findings indicate that integrating both acoustic and linguistic features leads to a noticeable improvement in prediction performance. The proposed hybrid model attains an accuracy of 76.44
Recent studies indicate that speech patterns and language usage can provide early indications of cognitive decline. Individuals affected by Alzheimer’s disease frequently show variations in speech fluency, vocabulary diversity, pause pat-terns, and sentence formation. These variations can be exam-ined using computational methods along with machine learning techniques.
In this work, we present a machine learning-based approach for identifying Alzheimer’s disease through speech recordings and linguistic analysis. Acoustic features such as Mel Frequency Cepstral Coefficients (MFCC), pitch, energy, speech rate, and pause duration are derived from audio signals. Linguistic features are obtained from transcribed speech using TF-IDF representation.
Multiple models, including text-based, audio-based, and combined feature models, are tested using the XGBoost classifier. The experimental findings indicate that integrating both acoustic and linguistic features leads to a noticeable improvement in prediction performance. The proposed hybrid model attains an accuracy of 76.44
Abstract
Alzheimer’s Disease Detection Using AI
Divyanshi Kashyap, Divya pandey, Mansi Pandey, Divyanshi Yadav, Dr. Nikhat Akhtar
DOI: 10.17148/IJARCCE.2026.15537
Abstract: Alzheimer’s disease is a progressive neurological condition that impacts memory, thinking ability, and speech functions. Early detection of Alzheimer’s is very crucial because timely intervention can help slow the progression of symptoms and enhance the overall quality of life for patients. Conventional diagnostic methods mainly depend on neuroimaging, cognitive assessments, and clinical evaluations, which are often costly, time-consuming, and not easily accessible to all individuals.
Recent studies indicate that speech patterns and language usage can provide early indications of cognitive decline. Individuals affected by Alzheimer’s disease frequently show variations in speech fluency, vocabulary diversity, pause pat-terns, and sentence formation. These variations can be exam-ined using computational methods along with machine learning techniques.
In this work, we present a machine learning-based approach for identifying Alzheimer’s disease through speech recordings and linguistic analysis. Acoustic features such as Mel Frequency Cepstral Coefficients (MFCC), pitch, energy, speech rate, and pause duration are derived from audio signals. Linguistic features are obtained from transcribed speech using TF-IDF representation.
Multiple models, including text-based, audio-based, and combined feature models, are tested using the XGBoost classifier. The experimental findings indicate that integrating both acoustic and linguistic features leads to a noticeable improvement in prediction performance. The proposed hybrid model attains an accuracy of 76.44
1
Recent studies indicate that speech patterns and language usage can provide early indications of cognitive decline. Individuals affected by Alzheimer’s disease frequently show variations in speech fluency, vocabulary diversity, pause pat-terns, and sentence formation. These variations can be exam-ined using computational methods along with machine learning techniques.
In this work, we present a machine learning-based approach for identifying Alzheimer’s disease through speech recordings and linguistic analysis. Acoustic features such as Mel Frequency Cepstral Coefficients (MFCC), pitch, energy, speech rate, and pause duration are derived from audio signals. Linguistic features are obtained from transcribed speech using TF-IDF representation.
Multiple models, including text-based, audio-based, and combined feature models, are tested using the XGBoost classifier. The experimental findings indicate that integrating both acoustic and linguistic features leads to a noticeable improvement in prediction performance. The proposed hybrid model attains an accuracy of 76.44
1
Abstract
Student Health and Stress Prediction System using Machine Learning
A Revanth, G J Sachidananda, G Mahesh Gouda, Dr. Muhibur Rahman T.R
DOI: 10.17148/IJARCCE.2026.15538
Abstract: This study introduces a Student Health and Stress Prediction System that uses machine learning to better understand students’ well-being. It looks at everyday habits like sleep, study time, screen usage, and physical activity to identify patterns related to stress. The data is collected through surveys and carefully processed to ensure it is accurate and useful. By applying machine learning techniques, the system can categorize students into different stress levels such as low, moderate, or high. Statistical methods are used to check how reliable the predictions are. The system also helps in detecting early signs of stress and suggests simple ways to improve daily routines. Overall, this approach supports students in maintaining better mental health, leading to improved well-being and academic performance.
Keywords: Student stress detection, machine learning techniques, monitoring of mental health, analysis of daily lifestyle habits, data-driven insights.
Keywords: Student stress detection, machine learning techniques, monitoring of mental health, analysis of daily lifestyle habits, data-driven insights.
Abstract
UNIEVENTS : A COLLEGE EVENT HANDLER USING AI
Laxmi Kanaujiya, Tanya Prajapati, Samriddhi Malhotra, Ranjana Kumari Chirania, Ms Prachi Yadav
DOI: 10.17148/IJARCCE.2026.15539
Abstract: This research presents UniEvents, a comprehensive and intelligent web-based platform designed to simplify and enhance the management of university events. In modern educational institutions, events such as technical workshops, seminars, cultural festivals, and competitions play a vital role in student development. However, the management of these events often relies on fragmented communication channels such as notice boards, social media platforms, and messaging applications, which leads to inefficiencies, lack of coordination, and reduced student participation. These challenges highlight the need for a centralized and automated solution.
The proposed system, UniEvents, addresses these issues by providing a unified platform where students and administrators can interact efficiently. The system enables students to browse available events, view detailed information, register online, and receive real-time notifications regarding updates and reminders. For administrators, the platform offers tools to create, modify, and manage events, monitor registrations, and maintain organized records of participants. This reduces manual effort, minimizes errors, and improves overall event coordination.
A key feature of the system is the integration of an AI-based recommendation mechanism, which analyzes user preferences, browsing behavior, and past participation to suggest relevant events. This personalization enhances user engagement and encourages increased participation. Additionally, the platform incorporates a chatbot system that provides instant responses to user queries, assists with navigation, and offers guidance regarding event-related information, thereby improving user experience and accessibility.
The system is developed using the MERN stack, which includes MongoDB for database management, Express.js and Node.js for backend processing, and React.js for the frontend interface. This technology stack ensures scalability, flexibility, and efficient handling of large volumes of data. Security measures such as user authentication and data protection are implemented to ensure reliability and privacy.
The results of the implementation demonstrate that UniEvents significantly improves the efficiency of event management processes within educational institutions. It enhances communication, increases student participation, and provides a structured approach to handling events. Overall, the proposed system offers a modern, intelligent, and scalable solution that bridges the gap between traditional event management methods and current technological advancements.
The proposed system, UniEvents, addresses these issues by providing a unified platform where students and administrators can interact efficiently. The system enables students to browse available events, view detailed information, register online, and receive real-time notifications regarding updates and reminders. For administrators, the platform offers tools to create, modify, and manage events, monitor registrations, and maintain organized records of participants. This reduces manual effort, minimizes errors, and improves overall event coordination.
A key feature of the system is the integration of an AI-based recommendation mechanism, which analyzes user preferences, browsing behavior, and past participation to suggest relevant events. This personalization enhances user engagement and encourages increased participation. Additionally, the platform incorporates a chatbot system that provides instant responses to user queries, assists with navigation, and offers guidance regarding event-related information, thereby improving user experience and accessibility.
The system is developed using the MERN stack, which includes MongoDB for database management, Express.js and Node.js for backend processing, and React.js for the frontend interface. This technology stack ensures scalability, flexibility, and efficient handling of large volumes of data. Security measures such as user authentication and data protection are implemented to ensure reliability and privacy.
The results of the implementation demonstrate that UniEvents significantly improves the efficiency of event management processes within educational institutions. It enhances communication, increases student participation, and provides a structured approach to handling events. Overall, the proposed system offers a modern, intelligent, and scalable solution that bridges the gap between traditional event management methods and current technological advancements.
Abstract
AI-Based Smart Home Intrusion Detection & Alert System Using Behavior Analysis and Face Recognition
Mallappa H, Yashwanth T M, Naveendra Reddy, Ameer S, Asst. Prof. Rajashekar Reddy P, Dr. Anita Patil
DOI: 10.17148/IJARCCE.2026.15540
Abstract: Home security has become a critical concern due to the increasing number of intrusion and theft incidents in residential areas. Traditional surveillance systems rely on manual monitoring and lack intelligent threat detection capabilities. This paper proposes an AI-based smart home intrusion detection and alert system that integrates face recognition and behavior analysis for continuous monitoring. The system captures real-time video data and processes it using deep learning algorithms to identify authorized and unauthorized individuals. It further analyzes human activity patterns to detect abnormal or suspicious behavior. Upon detecting an intrusion, instant alerts are sent to the homeowner through a mobile application. The proposed system enhances security by reducing response time and minimizing false alarms. It supports multi-modal inputs and ensures scalable deployment using IoT devices. The integration of artificial intelligence improves accuracy and automation in home surveillance. This system provides a reliable and efficient solution for modern smart home security.
Keywords: Smart Home Security, Intrusion Detection System, Artificial Intelligence, Face Recognition, Behavior Analysis, Computer Vision, Deep Learning, Continuous Monitoring, IoT-Based Surveillance, Real-Time Alert System
Keywords: Smart Home Security, Intrusion Detection System, Artificial Intelligence, Face Recognition, Behavior Analysis, Computer Vision, Deep Learning, Continuous Monitoring, IoT-Based Surveillance, Real-Time Alert System
Abstract
Generative AI: A Comprehensive Survey on Transformer-Based Models
J Hemanth, Harsha B, Ashish Kumar, Anurag N, Dr. Muhibur Rahman T.R
DOI: 10.17148/IJARCCE.2026.15541
Abstract: The landscape of artificial intelligence has undergone a profound transformation with the emergence of generative models capable of producing coherent text, realistic images, synthesized audio, and functional code. Central to this shift is the Transformer architecture, introduced by Vaswani et al. in their landmark 2017 contribution “Attention Is All You Need,” which fundamentally redefined how sequential data is modeled by replacing recurrence with parallelizable, attention-based processing [1]. This survey provides a structured and thorough examination of Transformer-based generative AI systems, tracing their development from early recurrent sequence architectures through to the most advanced multimodal and agentic AI frameworks. A four-tier taxonomic model is introduced to systematically classify these systems: basic recurrent sequence models, Transformer-based NLP architectures, large-scale language models, and multimodal autonomous AI agents. Comparative analysis is conducted across these tiers, evaluating performance, scalability, and contextual modeling capability. Through review of more than a dozen pivotal studies— spanning bidirectional pretraining via BERT [2], autoregressive generation through the GPT series [3][4][5], and visual representation learning via Vision Transformers [6]—this paper maps out the major inflection points in the field’s evolution. Critical open challenges are also examined, including the substantial computational overhead of large-scale training, the persistent problem of model hallucination, opacity in decision-making, systemic data biases, and the inability to adapt to real-time information. The paper concludes with a forward-looking perspective on next-generation generative AI, emphasizing efficiency, trustworthiness, and ethical design.
Keywords: Generative AI, Transformer Architecture, Self-Attention, BERT, GPT, Large Language Models, Vision Transformer, Multimodal AI, Natural Language Processing, Deep Learning.
Keywords: Generative AI, Transformer Architecture, Self-Attention, BERT, GPT, Large Language Models, Vision Transformer, Multimodal AI, Natural Language Processing, Deep Learning.
Abstract
AI BNB: AI-BASED TRAVEL LISTING AND BOOKING PLATFORM
Gaurav Yadav, Tanveer Alam, Balbir Singh, Rishabh Shukla, Ms. Deepika Pandey
DOI: 10.17148/IJARCCE.2026.15542
Abstract: This paper presents AI BnB, an AI-powered travel and accommodation booking platform built using the MERN stack (MongoDB, Express.js, React.js, Node.js). The platform addresses the growing need for intelligent, personalized property discovery by integrating a machine learning-based recommendation system that analyses user preferences, search history, and behavioural patterns. The system enables users to search, filter, and book properties seamlessly, while hosts manage listings and administrators oversee the platform. Key features include role-based JWT authentication, AI-driven property recommendations, smart search filters, secure booking management, and an AI Concierge powered by the Gemini API for natural language hotel queries. The platform is designed for scalability, responsiveness, and data security, demonstrating how AI combined with modern web development can significantly improve user experience in the hospitality domain.
Keywords: AI Recommendation System, Travel Booking, MERN Stack, Machine Learning, Property Listing, JWT Authentication, Gemini API, MongoDB
Keywords: AI Recommendation System, Travel Booking, MERN Stack, Machine Learning, Property Listing, JWT Authentication, Gemini API, MongoDB
Abstract
Smart Career Platform: An AI-Driven Personalised Career Guidance System for Multi-Domain Professionals Using MERN Stack and Groq Large Language Models
Purshottam Mishra, Awadhesh Kumar, Mr. Dileep Kumar Gupta
DOI: 10.17148/IJARCCE.2026.15543
Abstract: The proliferation of digital career platforms has disproportionately focused on technology professionals, leaving practitioners in law, medicine, finance, design, marketing, and education without adequate AI-driven career guidance. This paper presents the Smart Career Platform — a comprehensive full-stack career development system built on the MERN stack (MongoDB, Express.js, React.js, Node.js) and integrated with Groq's Llama 3.1 Large Language Model (LLM). The platform introduces a novel field-first personalisation architecture that propagates rich user context across all career guidance features: adaptive skill assessments, AI-generated career roadmaps, ATS-optimised resume building, AI mock interview practice, and a multi-session advisory chatbot. A four-step onboarding mechanism captures five context dimensions (field, user type, career goal, target role, and experience level) and injects this context into every AI prompt via system prompt engineering. The assessment engine employs a hybrid strategy: a curated static question bank for 12 established domains and dynamic AI-generated questions for unlimited novel domains. Deployed in production on Vercel and Render with MongoDB Atlas, the platform supports 10+ professional fields and 22+ career roles, achieving sub-4-second AI response times. Empirical comparative analysis demonstrates significant advantages over five leading career platforms across seven key capability dimensions. The system validates that prompt-engineered context injection is a cost-effective alternative to domain fine-tuning for specialised AI advisory applications.
Keywords: Artificial Intelligence, Career Guidance, MERN Stack, Groq AI, Llama 3.1, Large Language Model, Personalised Learning, ATS Resume Builder, Skill Assessment, React.js, MongoDB, Field-First Personalisation, Full- Stack Development.
Keywords: Artificial Intelligence, Career Guidance, MERN Stack, Groq AI, Llama 3.1, Large Language Model, Personalised Learning, ATS Resume Builder, Skill Assessment, React.js, MongoDB, Field-First Personalisation, Full- Stack Development.
Abstract
A STUDY OF FINANCIAL STATEMENT ANALYSIS OF HDFC BANK
Rushikesh Prakash Kapate, Prasad Shivajirao Kadam, Dr. Pradyuman Shastri
DOI: 10.17148/IJARCCE.2026.15544
Abstract: This research paper evaluates the financial performance of HDFC Bank through a comprehensive analysis of its financial statements. The study employs ratio analysis, trend analysis, and comparative analysis to assess key financial parameters, including profitability, liquidity, solvency, and efficiency. Secondary data has been collected from annual reports and financial disclosures of the bank.
The analysis indicates that HDFC Bank has achieved steady growth in revenue and profitability, supported by strong asset quality and efficient management practices. The bank maintains adequate liquidity and capital strength, reflecting financial stability and resilience. Despite facing industry competition and evolving regulatory requirements, the bank continues to perform consistently.
The study concludes that HDFC Bank holds a strong financial position in the Indian banking industry and offers reliable performance indicators for stakeholders. The findings contribute to a better understanding of financial statement analysis in the banking sector.
Keywords: Financial Statement Analysis, Profitability, Liquidity, Solvency, Banking Sector, Ratio Analysis, HDFC Bank
The analysis indicates that HDFC Bank has achieved steady growth in revenue and profitability, supported by strong asset quality and efficient management practices. The bank maintains adequate liquidity and capital strength, reflecting financial stability and resilience. Despite facing industry competition and evolving regulatory requirements, the bank continues to perform consistently.
The study concludes that HDFC Bank holds a strong financial position in the Indian banking industry and offers reliable performance indicators for stakeholders. The findings contribute to a better understanding of financial statement analysis in the banking sector.
Keywords: Financial Statement Analysis, Profitability, Liquidity, Solvency, Banking Sector, Ratio Analysis, HDFC Bank
Abstract
ZIGBEE BASED SMART HELMET FOR MINERS
Payal Survase, Pratik Ghortale, Ashwini Adasre
DOI: 10.17148/IJARCCE.2026.15545
Abstract: Mining is one of the most hazardous industries, exposing workers to dangerous gases, extreme temperatures, and unsafe working conditions. Traditional safety systems lack real-time monitoring and communication, leading to delayed responses in emergencies. This paper presents a ZigBee-based smart helmet designed for miners, integrating environmental and health monitoring sensors such as gas sensors (MQ-7, MQ-2), temperature sensor (DHT11), and pulse sensor (MAX30100). The system continuously monitors the miner’s condition and transmits real-time data using ZigBee wireless communication to a control station. In case of abnormal conditions, alerts are generated through buzzers and LEDs. This system enhances miner safety, reduces accident risks, and improves emergency response efficiency.
Keywords: Smart Helmet, ZigBee, Mining Safety, Gas Detection, IoT, Wearable Technology.
Keywords: Smart Helmet, ZigBee, Mining Safety, Gas Detection, IoT, Wearable Technology.
Abstract
Blockchain-Based Secure Voting System
Palak S Sachar, Sania Shaikh, Muktha Reddy, Dr. Muhibur Rehman T.R
DOI: 10.17148/IJARCCE.2026.15546
Abstract: The process of voting plays a crucial role in maintaining a fair and democratic system, yet existing methods often face issues related to trust, transparency, and security. Traditional paper-based voting can be slow and prone to manual errors, while electronic voting systems, though faster, still depend on centralized control, making them vulnerable to tampering and cyber threats. These challenges highlight the need for a more reliable and secure approach to conducting elections.
To address these concerns, this study proposes a blockchain-based voting system that ensures data integrity and transparency. By using a decentralized ledger, each vote is securely recorded and cannot be altered once it is added to the system. The use of encryption techniques helps maintain voter privacy, while the transparent nature of blockchain allows verification without exposing sensitive information. This approach minimizes the risk of fraud and increases confidence in the voting process.
The proposed system aims to create a balance between security, transparency, and usability. It provides a structured framework where voters can cast their votes securely and verify them if needed, without compromising anonymity. Overall, the system demonstrates how blockchain technology can be effectively applied to modernize voting systems and improve trust in digital elections.
Keywords: Blockchain, E-Voting, Security, Transparency, Smart Contracts, Decentralization
To address these concerns, this study proposes a blockchain-based voting system that ensures data integrity and transparency. By using a decentralized ledger, each vote is securely recorded and cannot be altered once it is added to the system. The use of encryption techniques helps maintain voter privacy, while the transparent nature of blockchain allows verification without exposing sensitive information. This approach minimizes the risk of fraud and increases confidence in the voting process.
The proposed system aims to create a balance between security, transparency, and usability. It provides a structured framework where voters can cast their votes securely and verify them if needed, without compromising anonymity. Overall, the system demonstrates how blockchain technology can be effectively applied to modernize voting systems and improve trust in digital elections.
Keywords: Blockchain, E-Voting, Security, Transparency, Smart Contracts, Decentralization
Abstract
Smart Hospital Management System
Rakesh Yadav, Arvind Yadav, Bhanu Chandra
DOI: 10.17148/IJARCCE.2026.15547
Abstract: This paper details the design, development, and features of a web-based Hospital Management System (HMS) aimed at streamlining patient registration, appointment scheduling, and administrative oversight. The system utilizes a three-module architecture (Patient, Doctor, Admin) providing distinct interfaces and functionalities for each user type. Built using HTML, CSS, JavaScript, and Bootstrap for the front-end, and PHP with a MySQL RDBMS for the back-end logic and data storage, the system offers features including user registration/login, appointment booking and cancellation, appointment history viewing, doctor management (addition/removal by admin), patient/doctor searching, and feedback viewing. The system offers an easy interface and demonstrates a real-world use of standard web technologies for managing basic healthcare needs. We have implemented this project in project-based learning.
Keywords: Hospital Management System, HMS, Web Application, PHP, MySQL, Bootstrap, Appointment Scheduling, Patient Management, Healthcare IT, Modular Design, Database Management
Keywords: Hospital Management System, HMS, Web Application, PHP, MySQL, Bootstrap, Appointment Scheduling, Patient Management, Healthcare IT, Modular Design, Database Management
Abstract
AFFECTIVE COMPUTING TECHNIQUES FOR HUMAN–MACHINE INTERACTION
Rabiya Fathima, C S Swetha
DOI: 10.17148/IJARCCE.2026.15548
Abstract: Affective computing is an emerging domain of artificial intelligence that focuses on enabling machines to recognize, interpret, and respond to human emotions. With the rapid advancement of intelligent systems such as chatbots, virtual assistants, and service robots, the ability to understand human emotions has become essential for improving interaction quality and user satisfaction. This paper presents a comprehensive survey and comparative analysis of various emotion recognition techniques, including facial expression recognition, speech emotion recognition, and text-based sentiment analysis. Deep learning models such as Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and Long Short-Term Memory (LSTM) networks are widely used for emotion detection tasks. Furthermore, multimodal approaches that combine multiple data sources are analyzed for their effectiveness in reducing ambiguity and improving accuracy. The study also discusses key challenges such as real-time processing, data bias, and privacy concerns. Finally, the paper highlights future research directions for developing efficient and human-centric emotion- aware systems.
Keywords: Affective Computing, Emotion Recognition, Deep Learning, CNN, LSTM, Multimodal Systems, Human– Machine Interaction
Keywords: Affective Computing, Emotion Recognition, Deep Learning, CNN, LSTM, Multimodal Systems, Human– Machine Interaction
Abstract
SafeHer: A Proactive AI-Driven Safety System for Women Using Contextual Risk Assessment and Real-Time Monitoring
Rohit Kumar Yadav, Aditya Upadhyay, Kajal Kasaudhan, Dr. Nikhat Akhtar
DOI: 10.17148/IJARCCE.2026.15549
Abstract: Women’s safety remains a critical global challenge, particularly in urban and semi-urban environments where conventional manual panic-button applications fail precisely when they are needed most. This paper presents SafeHer, a proactive, AI-driven mobile safety platform that shifts personal protection from reactive alerting to predictive, autonomous risk assessment. The system continuously monitors a user’s contextual environment—GPS location, motion patterns, time-of-day, network signal strength, and environmental isolation—and processes these signals through a two- stage Hybrid Danger Score Engine. The first stage applies deterministic rule-based thresholds on-device (Phase 1), producing a Base Risk Score (0–50) within milliseconds and without network dependency. The second stage transmits a structured context vector to Google Gemini AI via a secure Firebase Cloud Function proxy, obtaining an AI Risk Score (0–50) with contextual reasoning and one-sentence justification. The combined Final Danger Score (0–100) drives a threshold-triggered autonomous alerting pipeline that dispatches Firebase Cloud Messaging push notifications and Twilio SMS to pre-registered trusted contacts, shares a live GPS tracking link, and logs incident context—all without requiring any user interaction. Beyond automated alerting, SafeHer includes a Fake Call module for discreet social exit, an Evidence Vault with a system-update decoy screen for covert audio recording, a community Fear Map with differential privacy (Laplace noise, ε = 0.1) for anonymized urban safety analytics, and a Trust Score system for crowd-validating community hazard reports. Built on a Flutter frontend with a scalable Firebase backend, the platform achieves average alert latency of 1.8–2.7 seconds, background battery consumption of 9.3% over eight hours, a hybrid AUC of 0.94, and a System Usability Scale score of 82.4 across 30 simulated threat scenarios and a 15-participant pilot study.
Keywords: Women Safety, Predictive Safety Systems, AI-Based Risk Detection, Mobile Context Sensing, Firebase Architecture, Google Gemini AI, Proactive Alerting, Real-Time Location Tracking, Fake Call Feature, Evidence Vault, Fear Map, Trust Score, Differential Privacy.
Keywords: Women Safety, Predictive Safety Systems, AI-Based Risk Detection, Mobile Context Sensing, Firebase Architecture, Google Gemini AI, Proactive Alerting, Real-Time Location Tracking, Fake Call Feature, Evidence Vault, Fear Map, Trust Score, Differential Privacy.
Abstract
COVID SAFETY SYSTEM – DOOR HANDLE SANITIZER & TEMPERATURE DETECTION
Bhalghare Kajal Sanjay
DOI: 10.17148/IJARCCE.2026.15550
Abstract: Since December 2019 the world is under tremendous tension, the numbers are increasing day by day, and till date no vaccine has been full proved against the pandemic agent. The COVID-19 virus was unknown to us before it cast its outbreak in Wuhan, China. Being from a large family, a continuous mutation is occurring, forbidding the researchers, microbiologist, pharmaceuticals to search for the cure for the vaccine. Affecting the countries in a chain; China, Italy, Spain, USA, India , Russia, the virus has proved it’s strength and subservient a technologically enhanced race. The policies taken worldwide has lessen its affect to some extent but could not eradicate it. Lockdown has economically weaken many nations, and testing of different medicines has also not proven to be satisfactory. The design shows the preventive measure that can be taken during the COVID-19 pandemic in the whole world. Sanitizers have become the most significant commodities right now. By the new rules and regulations given by WHO vigorous sanitization is needed to survive. The design gave the solution for the problem stated. The design introduces an automatic hand sanitizer and temperature sensing system, to keep the hand sanitized whenever a person wants to do it, without a contact with the sanitizing machine. The temperature sensor on touching gives the body temperature of the person.
Keywords: Automatic hand sanitizer, Arduino, ultrasonic sensor, PIR sensors, TMP36, covid-19
Keywords: Automatic hand sanitizer, Arduino, ultrasonic sensor, PIR sensors, TMP36, covid-19
Abstract
LABOUR LAW COMPLIANCE AND ITS IMPACT ON EMPLOYEE RELATIONS IN ORGANIZATIONS
Shubhada Kumbhar
DOI: 10.17148/IJARCCE.2026.15551
Abstract: Labour law compliance plays a crucial role in maintaining fair and ethical workplace practices. This study examines the impact of labour law compliance on employee relations within organizations. Labour laws are designed to protect employee rights, ensure fair wages, safe working conditions, and promote harmonious industrial relations.
Non-compliance with labour laws can lead to disputes, dissatisfaction, legal penalties, and poor employee morale. On the other hand, effective compliance fosters trust, transparency, and positive relationships between employers and employees.
The study highlights how adherence to labour regulations improves employee satisfaction, reduces conflicts, and enhances organizational productivity. It also emphasizes the importance of HR policies, compliance mechanisms, and ethical leadership in strengthening employee relations.
Keywords: Labour Law Compliance, Employee Relations, Industrial Relations, Workplace Ethics, Job Satisfaction, HR
Non-compliance with labour laws can lead to disputes, dissatisfaction, legal penalties, and poor employee morale. On the other hand, effective compliance fosters trust, transparency, and positive relationships between employers and employees.
The study highlights how adherence to labour regulations improves employee satisfaction, reduces conflicts, and enhances organizational productivity. It also emphasizes the importance of HR policies, compliance mechanisms, and ethical leadership in strengthening employee relations.
Keywords: Labour Law Compliance, Employee Relations, Industrial Relations, Workplace Ethics, Job Satisfaction, HR
Abstract
AI-Powered Document Question Answering System Using Agentic RAG and Local Language Models
Varshitha D, Vidya S
DOI: 10.17148/IJARCCE.2026.15552
Abstract: The rapid growth of digital information and unstructured data has transformed the way users search, analyze, and interact with documents. However, traditional search systems often lack contextual understanding, intelligent reasoning, and the ability to generate precise answers from multiple sources. This paper presents DocAgent, an AI- powered Agentic Retrieval-Augmented Generation (RAG) system designed to provide accurate and context-aware responses from user-provided documents and web content. The proposed system integrates document ingestion, semantic search, and local large language model processing to enable efficient knowledge retrieval. Documents and URLs are processed into vector embeddings and stored in a vector database, allowing the system to perform similarity-based retrieval. An agent-based workflow dynamically decides whether to retrieve relevant context, refine the query, or directly generate responses, improving accuracy and reducing irrelevant outputs. The system utilizes a locally hosted language model through Ollama, ensuring privacy, cost efficiency, and reduced dependency on external APIs. The application is evaluated using real-time query scenarios, demonstrating improved response relevance, reduced hallucinations, and enhanced user interaction compared to traditional document search systems.
Keywords: Artificial Intelligence, Retrieval-Augmented Generation, Agentic AI, Document Question Answering, Ollama, Vector Database, Semantic Search, Web-Based System
Keywords: Artificial Intelligence, Retrieval-Augmented Generation, Agentic AI, Document Question Answering, Ollama, Vector Database, Semantic Search, Web-Based System
Abstract
Blockchain Driven Multi Tier Academic Credential Validation System
G. Priyadharshini M.E., Prakash B, Logesh P
DOI: 10.17148/IJARCCE.2026.15553
Abstract: The rise of fraudulent academic certificates has led to serious real-world issues, such as the use of fake degree certificates for jobs and visas. It also causes noteworthy disruptions in the hiring process. Although trust in institutions has been reduced, some research has been conducted to validate the authenticity of certificates. Blockchain- based verification systems focus on storing and validating certificates but often fail to provide integrated issuer authentication, tamper-proof privacy protection, and low-cost solutions. In this proposed work, research is done on an integrated, comprehensive, decentralised framework for certificate verification that together validates both the certificate and the issuer's authority. In addition, it protects the certificate from being tampered with. In addition, hash function mapping was employed for faster searching. The proposed solution is experimentally validated by creating a structure that incorporates zero-knowledge proofs (ZKPs) to safeguard data, self-destructing smart contracts to exclude replay attacks, and post-quantum lattice-based signatures to ensure security versus failures. In addition, a hash-based indexing method is used to accelerate certificate searches while reducing storage demands. In experimental analysis, measure the performance of proposed work in private. The Ethereum blockchain indicates that the proposed approach attains low transaction and gas costs, fast retrieval times, and high verification compared to previous research techniques. This overall analysis achieves 94% of a scalable and privacy-preserving solution for authenticating academic credentials in real-world scenarios.
Keywords: Block chain, Zero-knowledge proofs (ZKPs), Ethereum, Tamper-Proof, post-quantum lattice, hash-based indexing, smart contract.
Keywords: Block chain, Zero-knowledge proofs (ZKPs), Ethereum, Tamper-Proof, post-quantum lattice, hash-based indexing, smart contract.
Abstract
IndianAura: A Secure and Scalable E-Commerce Platform for Certified Made-in-India Products
Aishwarya V S, Anagha I V, Bindu B, Deepika G S, Veeramreddy Rajasekhar
DOI: 10.17148/IJARCCE.2026.15554
Abstract: The proliferation of counterfeit and non-certified goods on mainstream e-commerce platforms poses a significant threat to consumer trust and the growth of indigenous manufacturing ecosystems. Existing platforms such as Amazon and Flipkart lack structured mechanisms to verify product authen- ticity at the point of listing. This paper presents IndianAura, a secure, role-based, and scalable web e-commerce platform for certified Made-in-India products. The system introduces a multi-tier Product Certification Workflow with mandatory administrative approval. It enforces strict Role-Based Access Control (RBAC) across Administrator, Producer, and Customer roles. A Smart Scan feature leverages AI-based image analysis for product identification, while an Audit Logging subsystem ensures transparency and accountability. Built on a MERN-aligned architecture, IndianAura achieves a mean API response time of 187 ms under 500 concurrent users and a verification accuracy of 94.3%. Results demonstrate improved product trustworthiness, transparency, and certification compliance.
Keywords: E-commerce, Role-Based Access Control, Prod- uct Certification, Audit Logging, Image Recognition, Make in India, Secure Web Architecture.
Keywords: E-commerce, Role-Based Access Control, Prod- uct Certification, Audit Logging, Image Recognition, Make in India, Secure Web Architecture.
Abstract
A Secure Task Management System for Employee Workflow Optimization
G. Priyadharshini M.E., Priyadharshene R, Vimandhani G
DOI: 10.17148/IJARCCE.2026.15555
Abstract: A task management system for employees to work better is a website that helps people do their jobs. It is made to help people in the company work together in a way. This system can put all the tasks in one place so everyone can see what is going on. The tasks include finding security risks fixing problems checking what people are doing making sure only the right people can do things and dealing with incidents that happen during work. The task management system is, like a place where everything is organized and easy to find which helps the company security team do their job better. It is also hard to figure out who is responsible, for what task. The task management system provides role-based access. This means tasks are assigned to users based on their responsibility. Tasks are secure because authorized users can see and manage the tasks that are related to the tasks. The back end of the system uses PHP and MySQL. The system gives us a place to manage tasks create user accounts and store data for audits. It uses HTML5, CSS3 and JavaScript to make dashboards and interfaces that're easy to use. These interfaces help the user track progress check when things are due and get updates. The system lets the team manage tasks based on how important they're work. This means the team can work on the important tasks and fix problems first. The audit log is, like a record that shows what actions the user has taken. The user can check this record to see what they have done. This project is about making task management. It brings together ideas from task management, security operations and compliance tracking. The Secure task management project can help the team do tasks in a secure way. It also helps the team follow standards like the ISO/IEC 27001.The system is good at handling cybersecurity operations. The Secure task management project is useful for making the workflow platform better. It can help the team achieve 90% of the Secure task management projects goals. The Secure task management project is really good, for getting work done in a way. It may support the real-time task execution and monitoring of the audit tracking.
Keywords: Secure Task Management, Role-Based Access Control (RBAC), Cybersecurity Workflow, Audit Logging and Compliance, Web-Based Management System.
Keywords: Secure Task Management, Role-Based Access Control (RBAC), Cybersecurity Workflow, Audit Logging and Compliance, Web-Based Management System.
Abstract
AI-Enabled Eye Screening Tool for Early Detection of Common Eye Diseases
G. Priyadharshini M.E., Balananthakumar B, Jancy M
DOI: 10.17148/IJARCCE.2026.15556
Abstract: Eye diseases are common but they are often not find until later, especially in rural areas. It takes a long time and costs a lot of money to get a traditional diagnosis because you have to go to the hospital, use equipment, and see experts. Current AI-based systems are hard to use in real time because they are so complicated.
This project makes a basic Eye Disease Detection System using a CNN model. The user can either upload an eye image or take one through a web app. The system looks at the picture and finds diseases like cataracts, conjunctivitis, styes, or a normal condition.
Camera support, a better user interface, and better performance make the system better. It gives you quick and correct results along with a confidence score. The system is easy to use, works well, and is good for finding eye diseases early.
Keywords: Convolutional Neural Network (CNN), Machine Learning model Deep Learning, Early eye diseases detection, Accuracy, Artificial intelligence.
This project makes a basic Eye Disease Detection System using a CNN model. The user can either upload an eye image or take one through a web app. The system looks at the picture and finds diseases like cataracts, conjunctivitis, styes, or a normal condition.
Camera support, a better user interface, and better performance make the system better. It gives you quick and correct results along with a confidence score. The system is easy to use, works well, and is good for finding eye diseases early.
Keywords: Convolutional Neural Network (CNN), Machine Learning model Deep Learning, Early eye diseases detection, Accuracy, Artificial intelligence.
Abstract
Intelligent Taste Prediction Engine Using Scroll, Hover & View-Time Patterns in E-Commerce
G. Priyadharshini M.E., Roshani C, Pradeesh T
DOI: 10.17148/IJARCCE.2026.15557
Abstract: E-commerce platforms usually recommend products by looking at a user’s past activities, such as the items they clicked on, rated, or purchased earlier. While this works well for regular users, it does not clearly understand what a user wants during live browsing. This problem is more serious for first-time visitors and users who are not logged in, often resulting in general recommendations, low user engagement, and higher cart abandonment. Although advanced models like deep learning, graph-based methods, and transformers improve accuracy, they need large amounts of historical data, heavy computation, and offline training. Because of this, they are not practical for medium-scale e-commerce platforms. To solve these issues, this paper introduces a lightweight, intent-aware recommendation system that works in real time. The system observes small but meaningful user actions such as how far a user scrolls, how long they hover over products, how much time they spend viewing items, and how often they switch between products. Using these signals, the system understands the user’s intent during the same browsing session and updates recommendations instantly. It does not depend on personal or stored user data, which helps protect user privacy. Experimental results show that the proposed system achieves around 88–90% accuracy, reaching nearly 90% of the performance of complex existing models, while using much less computing power. This makes the system efficient, privacy-friendly, and suitable for real-time use in small and medium-scale e-commerce applications.
Keywords: E-commerce recommendation systems, session-based recommendation, real-time personalization, micro-interaction analysis, user intent prediction, lightweight machine learning, cold-start problem, privacy-preserving recommendation, small and medium enterprises (SMEs).
Keywords: E-commerce recommendation systems, session-based recommendation, real-time personalization, micro-interaction analysis, user intent prediction, lightweight machine learning, cold-start problem, privacy-preserving recommendation, small and medium enterprises (SMEs).
Abstract
Seasonal Crop Price Prediction using AI/ML
Aniket Sawant, Harsh Dudhal, Tejas Jadhav, Sarthak Kakade, Mrs. Dhage T.S.
DOI: 10.17148/IJARCCE.2026.15558
Abstract: Seasonal crop price prediction plays an important role in modern agriculture by helping farmers, traders, and government agencies make better economic decisions. Traditional forecasting methods often fail to capture complex seasonal patterns, weather variations, and market fluctuations present in agricultural data. This project proposes a crop price prediction system using Long Short-Term Memory (LSTM), a deep learning technique designed for time-series forecasting. The proposed model utilizes historical crop prices along with seasonal, climatic, and market-related factors such as rainfall, temperature, humidity, and previous market trends to predict future crop prices accurately. LSTM networks are highly effective in learning long-term dependencies and sequential patterns in time-series datasets, making them suitable for agricultural price forecasting
Keywords: Seasonal Crop Price Prediction, LSTM, Machine Learning, Deep Learning, Time-Series Forecasting, Smart Agriculture, Agricultural Data Analysis.
Keywords: Seasonal Crop Price Prediction, LSTM, Machine Learning, Deep Learning, Time-Series Forecasting, Smart Agriculture, Agricultural Data Analysis.
Abstract
HELMET PROTECTION DETECTION SYSTEM USING YOLOV8 FOR REAL-TIME TRAFFIC SAFETY MONITORING
Ramya Patani, Dr. Darapu Uma
DOI: 10.17148/IJARCCE.2026.15559
Abstract: Road traffic accidents involving two-wheelers result in significant fatalities, with head injuries being the primary cause. Traditional manual helmet compliance monitoring suffers from limited coverage, human error, and scalability issues. This paper presents an automated Helmet Protection Detection System using YOLOv8, a state-of-the-art real-time object detection model. The system processes multiple input sources including static images, video files, live webcam feeds, and uniquely, YouTube livestreams. The YOLOv8 model is trained to classify motorcyclists into two categories: "Helmet" (compliant) and "No- Helmet" (violator), with color-coded bounding boxes and confidence scores. Experimental results demonstrate high detection accuracy with an overall 92%, precision of 90%, and recall of 87%. The system achieves real-time performance of 45 FPS on a standard GPU, making it suitable for live traffic monitoring. This cost-effective solution leverages existing camera infrastructure, reduces dependency on manual supervision, and contributes to enhanced traffic safety enforcement.
Keywords: YOLOv8, Helmet Detection, Real-Time Monitoring, Deep Learning, Traffic Safety, Computer Vision.
Keywords: YOLOv8, Helmet Detection, Real-Time Monitoring, Deep Learning, Traffic Safety, Computer Vision.
Abstract
Phishing URL and Text Detection
R. Janaki M.E., (phd), Akalya A, Dharshini J
DOI: 10.17148/IJARCCE.2026.15560
Abstract: Phishing attacks are going up fast and they are a big problem when we are online. People who do these attacks use links to websites and fake messages to get important information from us, like our passwords and bank details and the special codes we get to confirm who we are. A lot of the systems we have can only find fake website links or fake messages not both so they do not work as well as they should. Phishing attacks use ways to trick us like fake website links and phishing messages to get our sensitive information. Previous studies used machine learning and deep learning techniques like Random Forest, Logistic Regression, Naïve Bayes, CNN, LSTM and BERT. These machine learning techniques were mainly used for detecting things in URLs. Text-based methods were also used, such as TF-IDF, Bag of Words and keyword analysis. The thing is, there is no simple system that can detect both URLs and messages at the same time, which is what machine learning and deep learning techniques, like Random Forest, Logistic Regression, Naïve Bayes, CNN, LSTM and BERT are supposed to do for URLs and messages. It uses machine learning and natural language processing to do this. The system looks at things like how long a website address how many dots it has and if it has special characters. It also checks if the website address uses HTTPS and if it is an IP address. The system uses tools like Logistic Regression and Random Forest to analyze all these things. It uses natural language processing to get the messages ready. Then it uses special tools like TF-IDF and Naïve Bayes with Logistic Regression. This helps the system figure out if the messages are trying to trick people into doing something or if they are regular messages. The phishing detection system is made to catch messages that try to scare people into doing something. The system is, about catching phishing and it uses machine learning and natural language processing to make sure it works well. The combined system improves detection accuracy, achieving 94% accuracy for phishing URLs and 92% accuracy for phishing text. The system can be extended for real-time use in browsers, emails, and SMS to protect users from online fraud.
Keywords: Phishing Detection, Phishing URL Detection, Phishing Text Detection, Machine Learning, Natural Language Processing (NLP), Cyber Security, Online Fraud Detection, Feature Extraction, Classification Models.
Keywords: Phishing Detection, Phishing URL Detection, Phishing Text Detection, Machine Learning, Natural Language Processing (NLP), Cyber Security, Online Fraud Detection, Feature Extraction, Classification Models.
Abstract
An Integrated IoT and Web-Based Framework for Energy and Machine Monitoring in SMEs
S Abhinaya, M.E., Akash M, Manimaran B
DOI: 10.17148/IJARCCE.2026.15561
Abstract: This project is about creating a system that helps industries use energy wisely and keep their machines in shape. The system is called a Smart IoT-Based Energy and Machine Condition Monitoring System. It is meant to make industries use energy efficiently make sure machines are reliable and find problems in time. The system uses a computer called an ESP32 microcontroller to control everything. This computer is connected to sensors. There is a PZEM-004T energy sensor that checks how much voltage, current, power and energy are being used. There is also an SW-420 vibration sensor that checks if machines are vibrating much. There is a DHT11 temperature sensor that checks how hot or cold it is around the machines. The Smart IoT-Based Energy and Machine Condition Monitoring System is very useful, for industries. It helps them use energy wisely and keep their machines in shape. The Smart IoT-Based Energy and Machine Condition Monitoring System is a way to make industries run better. The sensor data that we collect is looked at carefully. We use a method to decide what is normal and what is not. This method is like a warning system that checks the data all the time to make sure it is within limits. If something is wrong like the energy consumption is too high or the machine is vibrating much or getting too hot the system sends us a message right away. It also sends the data to the internet through Wi-Fi so we can look at it later. We use websites like Firebase to look at the data, in real time keep a record of it and see what happened in the past. The system we are talking about is set up from the edge to the cloud. This means it can do things quickly and does not get too busy with a lot of data. It is also cheap to set up. When we tried it out the system was very good at monitoring energy it got it right 95% of the time. It was also very good at finding faults using vibrations it got it right 90% of the time. And it was very good at monitoring temperature it got it right 92% of the time. The system is a help because it can reduce the amount of energy that is wasted by about 20 to 25%. It can also help find faults which is about 30% better than old ways of monitoring. It can reduce the amount of time machines are broken and cannot be used by about 40%. This is a lot better than the ways of monitoring machines.
Keywords: Industrial IoT, Energy Monitoring, Machine Monitoring, Web applications, SME Industry, Fault Detection, Industrial Sensors.
Keywords: Industrial IoT, Energy Monitoring, Machine Monitoring, Web applications, SME Industry, Fault Detection, Industrial Sensors.
Abstract
Detection of thyroid stages classification by convolutional neural network techniques
R. Janaki M.E.(phd), Ramya Shree V, Sweatha N
DOI: 10.17148/IJARCCE.2026.15562
Abstract: Thyroid disease diagnosis and stage classification are critical tasks in medical imaging due to their direct impact on patient treatment and management. Conventional diagnostic approaches based on manual interpretation of thyroid ultrasound images are time-consuming and prone to human error. To address these limitations, this paper presents an automated thyroid stage classification framework using Convolutional Neural Network (CNN) techniques.The proposed work is developed by taking an existing deep learning-based thyroid detection model as the base paper. Approximately 70% of the methodology is derived from the base paper, including image preprocessing concepts and deep feature extraction principles. The remaining 30% represents the proposed project contribution, where the system architecture is simplified and optimized using a CNN-focused approach for effective thyroid stage classification. The methodology involves preprocessing of thyroid ultrasound images to enhance image quality, followed by CNN-based automatic feature extraction and classification into different thyroid stages such as normal and abnormal. Deep learning technology is employed to eliminate manual feature engineering and improve classification performance. Experimental evaluation demonstrates that the proposed CNN-based model provides reliable accuracy and efficient classification compared to traditional diagnostic methods.The results indicate that CNN techniques are effective for thyroid stage classification and can be utilized as a supportive decision-making tool in clinical environments.
Keywords: Thyroid disease detection, thyroid stage classification, convolutional neural network (CNN), medical image classification, ultrasound image processing, deep learning.
Keywords: Thyroid disease detection, thyroid stage classification, convolutional neural network (CNN), medical image classification, ultrasound image processing, deep learning.
Abstract
PHISHING WEBSITE DETECTION SYSTEM USING ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING
Mr. D.S. Jaybhay, Ms. Chaitrali S. Shinde, Ms. Bhakti D. Nannaware, Ms. Sakshi A. Harnawal Ms. Priyanka S. Gadhe
DOI: 10.17148/IJARCCE.2026.15563
Abstract: Phishing websites are one of the most common cybersecurity threats used by attackers to steal sensitive information such as usernames, passwords, banking details, and personal data. These fake websites are designed to look similar to legitimate websites, making it difficult for users to identify them manually. Traditional phishing detection methods such as blacklist-based and rule-based systems are not effective for newly generated or unknown phishing websites.
This research presents a Phishing Website Detection System using Artificial Intelligence and Machine Learning. The proposed system analyzes different URL and website-based features such as URL length, number of special characters, suspicious keywords, domain-related attributes, login form presence, iframe usage, redirection behavior, and external links. Machine learning models such as Random Forest, XGBoost, and Multi-Layer Perceptron are used for classification. A hybrid ensemble voting model is applied to improve prediction accuracy and reliability.
The system is implemented as a Flask-based web application where users can enter a website URL and receive an instant prediction result. The output includes the classification result, phishing probability, risk level, important feature values, scan history, and downloadable PDF report. Experimental results show that the ensemble model performs better than individual classifiers and provides an effective solution for phishing website detection.
Keywords: Phishing Detection, Artificial Intelligence, Machine Learning, Cybersecurity, URL Analysis, Flask, Ensemble Learning, Website Security.
This research presents a Phishing Website Detection System using Artificial Intelligence and Machine Learning. The proposed system analyzes different URL and website-based features such as URL length, number of special characters, suspicious keywords, domain-related attributes, login form presence, iframe usage, redirection behavior, and external links. Machine learning models such as Random Forest, XGBoost, and Multi-Layer Perceptron are used for classification. A hybrid ensemble voting model is applied to improve prediction accuracy and reliability.
The system is implemented as a Flask-based web application where users can enter a website URL and receive an instant prediction result. The output includes the classification result, phishing probability, risk level, important feature values, scan history, and downloadable PDF report. Experimental results show that the ensemble model performs better than individual classifiers and provides an effective solution for phishing website detection.
Keywords: Phishing Detection, Artificial Intelligence, Machine Learning, Cybersecurity, URL Analysis, Flask, Ensemble Learning, Website Security.
Abstract
AI Driven IDS System in Network Security
Kartik P Shetty, Prof. Theerthashree G S
DOI: 10.17148/IJARCCE.2026.15564
Abstract: The increasing use of internet technologies, cloud computing, and smart devices has significantly increased cyber threats in modern networks. Traditional Intrusion Detection Systems (IDS) are unable to effectively detect advanced and unknown attacks because they rely mainly on predefined signatures and static security rules. Artificial Intelligence (AI) based IDS provides an intelligent and adaptive approach for improving network security. This paper presents a seminar-based study on AI Driven IDS Systems in Network Security using Machine Learning and Deep Learning techniques. The proposed system analyzes network traffic patterns, identifies abnormal behavior, and detects cyberattacks with improved accuracy and reduced false alarm rates. Various AI algorithms such as Random Forest, Support Vector Machine, Convolutional Neural Network, and Long Short-Term Memory are discussed in this paper. The study highlights the importance of AI-driven security systems in detecting both known and unknown threats efficiently. The proposed approach enhances overall network protection and provides better adaptability against evolving cyberattacks.
Keywords: Artificial Intelligence, Intrusion Detection System, Machine Learning, Deep Learning, Network Security, Cybersecurity
Keywords: Artificial Intelligence, Intrusion Detection System, Machine Learning, Deep Learning, Network Security, Cybersecurity
Abstract
Brain Tumor detection using Artificial Intelligence
Rushikesh Todekar, Shejal Kawale, Sakshi Khankar, Mayuri Sudake, Dr. Sachin Bere, Prof. Mrs. Jagtap P.S
DOI: 10.17148/IJARCCE.2026.15565
Abstract: Brain tumors are one of the most urgent forms of brain disease because they require early detection for effective treatment. The traditional method of diagnosing them is through MRI scans by expert Radiologists and is a time- consuming process. This contribution provides an Artificial Intelligence (AI) supported framework for the automatic detection and classification of brain tumors using Deep Learning. The NeuroScout platform is based on a ResNet based Convolutional Neural Network (CNN) algorithm trained on brain MRI images to classify brain tumors into four categories: Glioma, Meningioma, Pituitary Tumor and No Tumor. The NeuroScout platform contains FastAPI for the backend, Next.js for the frontend, and MongoDB for data storage, which implements a full stack medical web application. Google Gemini AI generates a medical explanation, treatment recommendation and prevention guidelines for brain tumors which have been detected by the NeuroScout Platform. A major advantage of this approach is that the NeuroScout system demonstrates a high degree of classification accuracy and provides a simple, intuitive user interface for patients, doctors and administrators alike. By providing an AI and Deep Learning based medical support system to healthcare professionals for early diagnosis, this approach will help provide greater access to AI based medical decision support systems.
Keywords: Brain Tumor Detection, Deep Learning, MRI Analysis, ResNet, Artificial Intelligence, Medical Image Processing, Machine Learning, Healthcare AI.
Keywords: Brain Tumor Detection, Deep Learning, MRI Analysis, ResNet, Artificial Intelligence, Medical Image Processing, Machine Learning, Healthcare AI.
Abstract
TimelimeX – Exploring the History using Gen AI
Dr. C N Shariff, Anjali, B Aishwarya, Bhavana Joshi, G Sahethi
DOI: 10.17148/IJARCCE.2026.15566
Abstract: TimelineX is an innovative web-based application designed to revolutionize the way history is learned, explored, and experienced. By integrating advanced Generative AI technologies, the system transforms traditional, static historical facts into dynamic, immersive storytelling experiences. Users can explore historical events through an interactive dashboard or by using a keyword-based search system, enabling quick and intuitive access to a vast range of historical moments. Once an event is selected, the AI generates rich and contextual narrations, converting textual descriptions into natural-sounding voice output to enhance engagement and accessibility. A key feature of TimelineX is its ability to generate AI-powered visual content, including images and short video clips that depict the selected historical event. This multimodal approach makes learning more interactive, memorable, and suitable for diverse learning preferences. In addition to narrative and visual enhancements, the platform introduces history- based games designed to reinforce learning through play. These gamified elements help users test their knowledge, encourage active participation, and make the overall experience more enjoyable—particularly for students. By bridging education, artificial intelligence, and interactive media, TimelineX transforms traditional learning into a modern digital experience. The platform provides a personalized and engaging environment for students, educators, self- learners, and history enthusiasts. Through its combination of AI-generated content, intuitive navigation, and educational games, TimelineX makes historical exploration more accessible, captivating, and meaningful. Overall, it demonstrates the potential of AI to reshape digital education by turning history into an engaging journey rather than a static archive of information.
Keywords: Generative AI, Interactive Learning, Natural Language Processing, Text-to-Speech, AI Image Generation, Gamified Education, Historical Visualization, Educational Technology.
Keywords: Generative AI, Interactive Learning, Natural Language Processing, Text-to-Speech, AI Image Generation, Gamified Education, Historical Visualization, Educational Technology.
Abstract
Secure Digital Identity Verification Using Blockchain
K Senbagam M.E., Kishore S, Divya K
DOI: 10.17148/IJARCCE.2026.15567
Abstract: In today’s digital era, safe identity authentication has become a significant concern because of increased Cyber Attacks, identity theft, and breaches.
The conventional digital identity solution is primarily reliant on a centralized database, which retains individuals’ information in one place. This is more prone to hacking, misusing, and unauthorized use. The conventional digital identity solution has limitations, which are addressed to a great extent by blockchain technology in digital identity authentication. the blockchain offers a decentralized and immutable platform on which the identity information can be recorded and verified in a secure manner without the need for any centralized authority. Every identity transaction is encrypted and recorded on multiple nodes in a manner that is extremely difficult to modify or counterfeit. The utilization of the blockchain technology enables individuals to control their identity information in a self-sovereign manner to share the necessary information only while maintaining privacy. The utilization of keys for encryption and the utilization of smart contracts also adds to the trust and authenticity pertaining to the verification process Secure digital identity verification using blockchain reduces fraud, improves privacy protection, and speeds up authentication processes across various sectors such as banking, healthcare, education, and government services. Overall, blockchain-based identity systems create a more secure, reliable, and user-centric approach to digital identity management, helping build trust in modern digital ecosystems.
Keywords: Secure Digital Identity, Blockchain Technology, Identity Authentication, Decentralized Identity, Self- Sovereign Identity (SSI), Privacy Preservation, Cryptographic Security, Smart Contracts, Identity Verification, Cybersecurity.
The conventional digital identity solution is primarily reliant on a centralized database, which retains individuals’ information in one place. This is more prone to hacking, misusing, and unauthorized use. The conventional digital identity solution has limitations, which are addressed to a great extent by blockchain technology in digital identity authentication. the blockchain offers a decentralized and immutable platform on which the identity information can be recorded and verified in a secure manner without the need for any centralized authority. Every identity transaction is encrypted and recorded on multiple nodes in a manner that is extremely difficult to modify or counterfeit. The utilization of the blockchain technology enables individuals to control their identity information in a self-sovereign manner to share the necessary information only while maintaining privacy. The utilization of keys for encryption and the utilization of smart contracts also adds to the trust and authenticity pertaining to the verification process Secure digital identity verification using blockchain reduces fraud, improves privacy protection, and speeds up authentication processes across various sectors such as banking, healthcare, education, and government services. Overall, blockchain-based identity systems create a more secure, reliable, and user-centric approach to digital identity management, helping build trust in modern digital ecosystems.
Keywords: Secure Digital Identity, Blockchain Technology, Identity Authentication, Decentralized Identity, Self- Sovereign Identity (SSI), Privacy Preservation, Cryptographic Security, Smart Contracts, Identity Verification, Cybersecurity.
Abstract
VisionGuard: Deep Learning–Based Weapon Detection Framework
K. Rajavadhani, Aswini. R. C & Vijayalakshmi. J
DOI: 10.17148/IJARCCE.2026.15568
Abstract: We have a lot of surveillance systems in places and private areas these days. This means we really need to have systems that can find threats right away. Usually people watch the cameras. We use simple rules to figure out what is going on. This does not work very well when there are a lot of people around and things get complicated. This paper is about a surveillance system that uses artificial intelligence and looks at pictures to find threats, in real time and understand what is happening. The system uses deep learning techniques to do this. The system they are talking about uses a Swin Transformer to find objects like weapons or unattended bags. It can also detect when someone is not supposed to be in an area. They also have a model that looks at how people move to see if they are doing something weird. This model uses something called Graph Convolutional Networks. It can tell if someone is just hanging around being violent or moving in a way. The Swin Transformer and the movement model work together to make sure the system is accurate and does not send out a lot of warnings. The system is really good at finding objects and strange movements, like the Swin Transformer finding weapons and the movement model finding unusual movement patterns. When we find something that could be a problem the system sends out alerts to help the security team act fast. The tests we did show that this way of doing things works well even in tough situations, which means it is a good choice for new smart surveillance systems that are being developed.
Keywords: Surveillance Systems, Threat Detection, Computer Vision, Deep Learning, Swin Transformer, Graph Convolutional Networks, Situational Awareness
Keywords: Surveillance Systems, Threat Detection, Computer Vision, Deep Learning, Swin Transformer, Graph Convolutional Networks, Situational Awareness
Abstract
Review on Potential Applications of Geopolymer Concrete in Construction
Vasanth S, Likhith Kumar T, Vamshi A, Mohammed Afiz Roshan
DOI: 10.17148/IJARCCE.2026.15569
Abstract: This paper review discusses the concept and applications of geopolymer concrete (GPC) as an alternative to conventional cement concrete. The reviewed study explains how industrial by-products such as fly ash and slag can be used to produce eco-friendly concrete. This review summarizes the main points of the paper, including materials used, mechanical properties, and practical uses. It also evaluates the strengths and weaknesses of the study. The findings show that geopolymer concrete is a promising material for sustainable construction, although more research is required for large-scale implementation.
Keywords: Geopolymer concrete, eco-friendly materials, fly ash, sustainable construction, alkali activation
Keywords: Geopolymer concrete, eco-friendly materials, fly ash, sustainable construction, alkali activation
Abstract
CallMind: An AI-Powered Real-Time Voice Agent Platform Using a Pluggable Speech-to-Text, Large Language Model, and Text-to-Speech Pipeline
Sahil Makandar, Rohini Magamdum, Aditya Mali, Sudesh Kumbhar, Radheshyam Sah, Rajendra Hiremath
DOI: 10.17148/IJARCCE.2026.15570
Abstract: Intelligent voice agents offer a scalable alternative to traditional Interactive Voice Response (IVR) systems and human call centres, yet their deployment remains technically complex. This paper presents CallMind, an AI-powered voice agent platform that enables businesses and individuals to deploy conversational telephone agents within minutes. The platform implements a pluggable, channel-agnostic pipeline that normalises all input modalities—phone calls, browser audio, and direct text—into a unified QueryPayload abstraction before processing. The core intelligence pipeline routes each query through Azure Cognitive Speech Services for real-time streaming Speech-to-Text (STT), Groq API (LLaMA 3.3 70B) for large language model inference, and Azure Neural Text-to-Speech (TTS) for audio synthesis. A sentence-by-sentence streaming architecture achieves end-to-end response latency of approximately 780 ms, comparable to natural human conversational response time. The system is built on a dual-backend architecture: a managed Supabase layer for multi-tenant agent configuration, knowledge base management, and conversation persistence; and a containerised Python FastAPI server on Microsoft Azure for the real-time AI pipeline. Live validation through Twilio Programmable Voice demonstrates natural conversation quality with accurate transcription, contextually relevant responses, and seamless audio delivery. The architecture provides a clear migration path from full-context LLM prompting to Retrieval-Augmented Generation (RAG) and from Docker Compose to Kubernetes without modifying the application layer.
Keywords: Voice Agent, Speech-to-Text, Text-to-Speech, Large Language Model, Real-Time Pipeline, WebSocket, Twilio, FastAPI, Docker, Pluggable Architecture, Conversational AI, Redis, Supabase, Retrieval- Augmented Generation
Keywords: Voice Agent, Speech-to-Text, Text-to-Speech, Large Language Model, Real-Time Pipeline, WebSocket, Twilio, FastAPI, Docker, Pluggable Architecture, Conversational AI, Redis, Supabase, Retrieval- Augmented Generation
Abstract
Smart Pothole Detection and Rapid Emergency Response System for Four-Wheeler Vehicles
Abhishek N, Dr. Kavitha A S, Nandish S, Sanjana Sindol, Bhavana
DOI: 10.17148/IJARCCE.2026.15571
Abstract: Potholes are one of the most dangerous road hazards in India, causing tyre bursts, loss of vehicle control, and fatal accidents every monsoon season. Yet no vehicle-mounted system currently warns the driver in real time and automatically notifies family members when a dangerous incident follows. This paper presents a Smart Pothole Detection and Rapid Emergency Response System for four-wheeler vehicles. A YOLOv8n model on a Raspberry Pi 4 detects potholes at 18 FPS from a dashboard camera, immediately alerting the driver via buzzer, LED, and mobile app. An MPU- 6050 inertial sensor then monitors vehicle dynamics for 10 seconds; if acceleration exceeds 4G or yaw rate surpasses 45°/s, a danger event is confirmed. A photo snapshot and live GPS coordinates are dispatched simultaneously via Firebase Cloud Messaging and GSM SMS to family members, whose app opens automatically showing the photo and location. Hardware costs under Rs. 9,000 with no vehicle modification required. Tests across Bengaluru roads confirmed 87.3% mAP detection and sub-5-second family alert delivery.
Keywords: Pothole Detection, YOLOv8, Emergency Response, IoT, Raspberry Pi, GPS Tracking, MPU-6050, Flutter Mobile App, Firebase, Road Safety, Four-wheeler Vehicles, Driver Alert System.
Keywords: Pothole Detection, YOLOv8, Emergency Response, IoT, Raspberry Pi, GPS Tracking, MPU-6050, Flutter Mobile App, Firebase, Road Safety, Four-wheeler Vehicles, Driver Alert System.
Abstract
Coder Buddy – An Agentic AI-Powered Python Development Assistant for Code Review, Code Generation, and Automated Mini Project Building
Sonika D, Harshitha L
DOI: 10.17148/IJARCCE.2026.15572
Abstract: The growing demand for intelligent programming assistants has led to the development of agentic AI systems that can autonomously plan and execute multi-step development tasks. This paper presents Coder Buddy, an agentic AI application designed to assist Python developers with three core capabilities: automated code review, prompt-based code generation, and complete mini project building. The system combines rule-based static analysis with Large Language Models (LLMs) to deliver accurate, context-aware feedback and generate production-ready Python code. A key feature of the system is its ability to auto-detect user intent and dynamically generate either CLI or GUI (Streamlit-based) projects with proper folder structure, dependency files, and runnable output. The application is built using Python and Streamlit, and maintains session-based history for all interactions. Evaluation across simulated development scenarios demonstrates the system's effectiveness in improving code quality, reducing development time, and supporting learning and rapid prototyping.
Keywords: Agentic AI, Code Review, Code Generation, Large Language Models, Python, Streamlit, AST Analysis, Mini Project Builder, Automated Software Development
Keywords: Agentic AI, Code Review, Code Generation, Large Language Models, Python, Streamlit, AST Analysis, Mini Project Builder, Automated Software Development
Abstract
MCP-Based Context-Aware System Monitoring and Threat Detection Agent
G. Priyadharshini, M.E., Balaji A, Vishnu S, Mohamed Noufal M
DOI: 10.17148/IJARCCE.2026.15573
Abstract: This paper presents an MCP-Based Context-Aware System Monitoring and Threat Detection Agent, an intelligent, real-time cybersecurity monitoring platform leveraging the Model Context Protocol (MCP) to deliver context-aware threat detection and automated response capabilities. Traditional Security Information and Event Management (SIEM) systems rely on static rule-based engines that produce false-positive rates of 25–45%, suffer from alert fatigue, and fail to detect multi-stage Advanced Persistent Threats (APTs). The proposed system integrates a FastAPI backend, PostgreSQL 16 storage, WireGuard VPN encryption, and a Bootstrap 5 web dashboard to provide unified, real-time visibility across network traffic, system logs, and behavioral metrics. The MCP AI agent maintains a rolling context window over incoming security events, enabling temporal correlation, multi-stage attack detection, lateral movement identification, and significant reduction of false positives through composite threat scoring. Validation results demonstrate a 67% reduction in false positives, sub-3-second automated mitigation response, throughput exceeding 6,200 concurrent events per second, an AUC of 0.94 on the ROC curve, and 60–75% reduction in operational costs versus commercial SIEM solutions. All three functional modules—Data Collection & Traffic Monitoring, Threat Analysis & Context Awareness, and Alerting & Secure Notification—have been implemented and validated in a prototype environment over a 30-day test period.
Keywords: Model Context Protocol (MCP), Cybersecurity, Real-Time Threat Detection, Context-Aware AI, SIEM, FastAPI, WireGuard VPN, Anomaly Detection, Automated Incident Response.
Keywords: Model Context Protocol (MCP), Cybersecurity, Real-Time Threat Detection, Context-Aware AI, SIEM, FastAPI, WireGuard VPN, Anomaly Detection, Automated Incident Response.
Abstract
AI-BASED VERIFICATION OF LLM RESPONSE
Mrs. Asha Sattigeri, Lavanya R, Pavan Gowda S
DOI: 10.17148/IJARCCE.2026.15574
Abstract: Large Language Models(LLMs) now handle tasks like question answering, summarisation, code generation and dialogue with impressive results. Yet they still suffer from a key issue: hallucination happens when a model generates text that reads well but is factually wrong or not grounded in real evidence. The risk is higher in domains like healthcare, law, finance and research, where inaccurate outputs can lead to real damage. This survey focuses on how to detect hallucination verification in LLMs. We review 5 core detection approaches: retrieval-based, uncertainty-based, embedding-based, learning-based and self-consistency methods. We also cover current mitigation techniques, popular benchmarks such as Truthful QA and HaluEval, common evaluation metrics and verification tools such as xVerify and CompassVerifier. This paper closes by discussing open challenges and future directions for building more reliable, truthful LLMs
Keywords: Large Language Models, Hallucination Detection, Hallucination Mitigation, Factuality, Retrieval- Augmented Generation, TruthfulQA, HaluEval, Answer verification, Claim verification, Internal states
Keywords: Large Language Models, Hallucination Detection, Hallucination Mitigation, Factuality, Retrieval- Augmented Generation, TruthfulQA, HaluEval, Answer verification, Claim verification, Internal states
Abstract
Controlling Ethical Hacking: Operating System Security
Dipali Girhe, R.V. Daund
DOI: 10.17148/IJARCCE.2026.15575
Abstract: The rapid advancement of technology has increased the risk of cyberAttacks, making ethical hacking a critical tool for securing information systems. Ethical hacking involves authorized testing of computer systems to identify vulnerabilities before malicious hackers exploit them. This research focuses on controlling ethical hacking through the use of operating systems such as Windows, Kali Linux, and Linux-based security distributions. Data is collected from secondary sources, including journals, research papers, official OS documentation, and cyber security reports, to analyse hacking tools, techniques, and OS-level security features. The study highlights how operating systems provide built-in protections, access controls, logging, auditing, and monitoring mechanisms that can prevent unauthorized access while enabling safe ethical hacking. Additionally, the research explores emerging trends and best practices for secure implementation of ethical hacking. The findings aim to provide a comprehensive understanding of how operating systems can strengthen cyber security defenses and control ethical hacking activities effectively.
Keywords: Ethical Hacking, Operating System, Security Features, Vulnerability Assessment, etc.
Keywords: Ethical Hacking, Operating System, Security Features, Vulnerability Assessment, etc.
Abstract
AI-Based Health Monitoring System
Ummi Habiba, Vijetha S P, C B Shekhara, Prajwal V, Anita Patil
DOI: 10.17148/IJARCCE.2026.15576
Abstract: Continuous and real-time health monitoring has emerged as one of the most pressing challenges in modern healthcare, particularly in the context of ageing populations, rising prevalence of chronic diseases, and the growing demand for remote patient care. This paper presents the design and implementation of an Artificial Intelligence (AI)- based health monitoring system capable of acquiring, processing, and analysing multiple physiological parameters — including heart rate, blood oxygen saturation (SpO2), body temperature, and blood pressure — in real time. The proposed system integrates low-cost wearable sensors with a microcontroller-driven edge computing unit and a cloud-connected dashboard, forming an end-to-end pipeline from data acquisition to clinical decision support. A hybrid machine learning model combining a Long Short-Term Memory (LSTM) network for temporal pattern recognition with a Random Forest classifier for anomaly labelling is trained on publicly available physiological datasets. The model achieves a classification accuracy of 94.7% in detecting critical health events such as tachycardia, hypoxia, and hypertensive episodes. Automated alert notifications are dispatched to caregivers and physicians whenever abnormal readings are detected, enabling timely intervention. Experimental evaluation demonstrates that the system maintains end-to-end latency below 1.8 seconds and sustains reliable operation across a 24-hour continuous monitoring window. The results confirm that integrating AI with wearable sensing technology offers a scalable and cost-effective approach to preventive healthcare.
Keywords: Artificial Intelligence, Health Monitoring, Wearable Sensors, LSTM, Random Forest, IoT in Healthcare, Remote Patient Monitoring, Anomaly Detection, Edge Computing, Physiological Signals.
Keywords: Artificial Intelligence, Health Monitoring, Wearable Sensors, LSTM, Random Forest, IoT in Healthcare, Remote Patient Monitoring, Anomaly Detection, Edge Computing, Physiological Signals.
Abstract
Iot Based Patient Health Monitoring System
Basavanagowda B S, Basavaraju Bailaannavar, Gagangowda, Vijaykumar Hovale, Prof. Padmavathi N*
DOI: 10.17148/IJARCCE.2026.15577
Abstract: The proposed system is an advanced Internet of Things (IoT)-based health monitoring solution designed to provide continuous and real-time tracking of a patient’s vital parameters.The system is built around the ESP32 microcontroller, which acts as the central processing unit and communication hub. It integrates multiple biomedical sensors, including a temperature sensor, heart rate sensor, ECG sensor, and blood pressure sensor, to collect accurate physiological data from the patient. The acquired data is processed by the ESP32 and displayed on an LCD screen for immediate local monitoring. In addition to local display, the system utilizes Wi-Fi connectivity to transmit the data to an IoT platform, enabling remote access through a caregiver’s smartphone. This allows doctors or caretakers to monitor thepatient’s health condition from any location in real time. The system also incorporates abuzzer for emergency alerts and a manual button that can be used by the patient to signal distress. These features enhance the safety and responsiveness of the system during criticalsituations. By combining sensor technology, wireless communication, and IoT capabilities,the proposed system offers a reliable, cost-effective, and efficient solution for remote health monitoring. It is particularly useful for elderly patients, individuals with chronic diseases, and situations where continuous medical supervision is required.
Abstract
Emotion-Aware Adaptive Learning Systems: A Comprehensive Survey on Artificial Intelligence-Based Personalized Education
A Rashmi, A Keerthi, Harika K, Likitha K, Dr. Muhibur Rahman T.R
DOI: 10.17148/IJARCCE.2026.15578
Abstract: The increasing adoption of digital learning platforms and intelligent educational technologies has transformed modern education systems. However, conventional e-learning environments often fail to adapt to the emotional and cognitive states of learners, resulting in reduced engagement, low motivation, and ineffective personalized learning experiences. To address these limitations, recent advancements in Artificial Intelligence (AI), affective computing, and adaptive learning systems have enabled the development of emotion-aware educational platforms capable of dynamically responding to student emotions and behavioral patterns. This survey presents a comprehensive review of Emotion-Aware Adaptive Learning Systems that utilize AI techniques for personalized education and intelligent learner interaction. The study systematically examines the evolution of affective computing in education, including emotion recognition methods based on facial expressions, speech analysis, physiological signals, eye tracking, and behavioral analytics. Furthermore, this survey analyzes the integration of Machine Learning, Deep Learning, Computer Vision, Natural Language Processing, and multimodal emotion recognition techniques in adaptive educational environments.
Through comparative analysis of recent studies, this paper evaluates the effectiveness of AI-driven adaptive learning systems in improving student engagement, learning efficiency, concentration, and academic performance. A four-layer taxonomy is proposed to classify existing systems into emotion detection, learner modeling, adaptive decision-making, and intelligent feedback mechanisms. The survey also highlights significant challenges including privacy concerns, emotional data reliability, ethical considerations, computational complexity, and real-time adaptability limitations. Additionally, emerging research trends such as explainable AI, multimodal affective computing, virtual intelligent tutors, and emotionally responsive educational agents are explored to identify future research directions. By consolidating existing research contributions and technological advancements, this survey aims to provide a structured understanding of emotion-aware AI learning systems and their potential to revolutionize personalized digital education. The findings of this study contribute toward the development of intelligent, human-centered educational technologies capable of enhancing learner experience and improving adaptive teaching methodologies in future smart learning environments.
Through comparative analysis of recent studies, this paper evaluates the effectiveness of AI-driven adaptive learning systems in improving student engagement, learning efficiency, concentration, and academic performance. A four-layer taxonomy is proposed to classify existing systems into emotion detection, learner modeling, adaptive decision-making, and intelligent feedback mechanisms. The survey also highlights significant challenges including privacy concerns, emotional data reliability, ethical considerations, computational complexity, and real-time adaptability limitations. Additionally, emerging research trends such as explainable AI, multimodal affective computing, virtual intelligent tutors, and emotionally responsive educational agents are explored to identify future research directions. By consolidating existing research contributions and technological advancements, this survey aims to provide a structured understanding of emotion-aware AI learning systems and their potential to revolutionize personalized digital education. The findings of this study contribute toward the development of intelligent, human-centered educational technologies capable of enhancing learner experience and improving adaptive teaching methodologies in future smart learning environments.
Abstract
Automated Threat Hunting Using AI: AI-Driven Defence Strategy
Vikram G.D, K. Sharath
DOI: 10.17148/IJARCCE.2026.15579
Abstract: The exponential digitization of global infrastructures, coupled with the proliferation of cloud computing, Internet of Things (IoT) devices, and decentralized workforces, has inadvertently created a vast and highly vulnerable cyber attack surface. Concurrently, adversarial tactics have evolved with alarming sophistication. Modern cybercriminals and state-sponsored actors frequently deploy zero-day exploits, advanced persistent threats (APTs), fileless malware, and polymorphic ransomware that are expressly designed to circumvent traditional, perimeter-based security architectures [1], [2]. Historically, cybersecurity has relied heavily on reactive paradigms—such as signature-based Intrusion Detection Systems (IDS) and standard firewalls—which require prior knowledge of an attack vector to mount a defense. This reactive posture is fundamentally insufficient in an era where the velocity and novelty of cyber threats outpace human response capabilities. To neutralize these stealthy incursions, the cybersecurity industry must pivot toward proactive threat hunting: the iterative, aggressive process of searching through networks, endpoints, and datasets to uncover latent malicious activities that have successfully evaded initial automated defenses.
However, the sheer volume and complex dimensionality of telemetry data generated by modern IT ecosystems render manual threat hunting physically impossible and highly susceptible to analyst burnout and alert fatigue. This paper presents a comprehensive framework for Automated Threat Hunting driven by Artificial Intelligence (AI) and Machine Learning (ML), positioning it as the indispensable core of modern cyber defense strategies [2], [3]. By integrating AI into Security Operations Centers (SOCs), organizations can transcend the limitations of human capacity. This research explores the deployment of advanced AI mechanisms, specifically focusing on User and Entity Behavior Analytics (UEBA) for establishing baseline operational norms, deep learning neural networks for structural payload analysis without relying on known signatures, and natural language processing (NLP) to autonomously ingest and correlate global threat intelligence feeds [6], [9].
Furthermore, this document examines how AI-driven systems leverage continuous contextual analysis to connect seemingly disparate, low-level alerts across vast network topologies, unearthing coordinated, slow-moving attacks before data exfiltration or encryption occurs. We also detail the integration of AI with Security Orchestration, Automation, and Response (SOAR) platforms to execute instantaneous, autonomous remediation protocols, drastically reducing both the Mean Time to Detect (MTTD) and Mean Time to Respond (MTTR) [10]. Finally, this paper critically assesses the practical implementation challenges—including high false-positive rates, data privacy constraints, and the emergence of adversarial AI [7], [11]—while forecasting the future scope of fully autonomous, self-healing networks governed by federated machine learning models [14].
Keywords: Automated Threat Hunting, Artificial Intelligence, Cybersecurity, Machine Learning, User and Entity Behavior Analytics (UEBA), Proactive Defense, Advanced Persistent Threats (APTs), Security Orchestration and Automation (SOAR), Anomaly Detection.
However, the sheer volume and complex dimensionality of telemetry data generated by modern IT ecosystems render manual threat hunting physically impossible and highly susceptible to analyst burnout and alert fatigue. This paper presents a comprehensive framework for Automated Threat Hunting driven by Artificial Intelligence (AI) and Machine Learning (ML), positioning it as the indispensable core of modern cyber defense strategies [2], [3]. By integrating AI into Security Operations Centers (SOCs), organizations can transcend the limitations of human capacity. This research explores the deployment of advanced AI mechanisms, specifically focusing on User and Entity Behavior Analytics (UEBA) for establishing baseline operational norms, deep learning neural networks for structural payload analysis without relying on known signatures, and natural language processing (NLP) to autonomously ingest and correlate global threat intelligence feeds [6], [9].
Furthermore, this document examines how AI-driven systems leverage continuous contextual analysis to connect seemingly disparate, low-level alerts across vast network topologies, unearthing coordinated, slow-moving attacks before data exfiltration or encryption occurs. We also detail the integration of AI with Security Orchestration, Automation, and Response (SOAR) platforms to execute instantaneous, autonomous remediation protocols, drastically reducing both the Mean Time to Detect (MTTD) and Mean Time to Respond (MTTR) [10]. Finally, this paper critically assesses the practical implementation challenges—including high false-positive rates, data privacy constraints, and the emergence of adversarial AI [7], [11]—while forecasting the future scope of fully autonomous, self-healing networks governed by federated machine learning models [14].
Keywords: Automated Threat Hunting, Artificial Intelligence, Cybersecurity, Machine Learning, User and Entity Behavior Analytics (UEBA), Proactive Defense, Advanced Persistent Threats (APTs), Security Orchestration and Automation (SOAR), Anomaly Detection.
Abstract
Medicinal Plant Identification and Classification
Anushree N N, Deepika V M
DOI: 10.17148/IJARCCE.2026.15580
Abstract: Medicinal plants play a significant role in healthcare systems due to their therapeutic and healing properties. Accurate identification of medicinal plants is essential for their proper utilization in traditional and modern medicine. However, manual identification requires expert botanical knowledge and is often time-consuming and error-prone because many plant species possess visually similar leaf structures. This project proposes an intelligent Medicinal Plant Identification and Classification system using deep learning techniques to automate the plant recognition process.
The proposed system utilizes Convolutional Neural Networks (CNNs) to analyze leaf images and classify medicinal plants based on visual features such as shape, texture, vein patterns, and color characteristics. Image preprocessing techniques including resizing, normalization, and augmentation are applied to improve model performance and handle variations in lighting, orientation, and background conditions. The system integrates Firebase Firestore for storing medicinal plant information such as scientific names, medicinal uses, and safety considerations.
Experimental evaluation demonstrates that the CNN-based approach provides high classification accuracy and reliable prediction results compared to conventional machine learning techniques. The proposed system reduces manual effort, improves identification speed, and provides an efficient and user-friendly solution for students, researchers, farmers, and healthcare enthusiasts. Overall, the project highlights the effective application of artificial intelligence and deep learning in botanical research and medicinal plant preservation.
Keywords: Medicinal Plants, Deep Learning, Convolutional Neural Network, Image Classification, Plant Identification, Artificial Intelligence, Firebase Firestore.
The proposed system utilizes Convolutional Neural Networks (CNNs) to analyze leaf images and classify medicinal plants based on visual features such as shape, texture, vein patterns, and color characteristics. Image preprocessing techniques including resizing, normalization, and augmentation are applied to improve model performance and handle variations in lighting, orientation, and background conditions. The system integrates Firebase Firestore for storing medicinal plant information such as scientific names, medicinal uses, and safety considerations.
Experimental evaluation demonstrates that the CNN-based approach provides high classification accuracy and reliable prediction results compared to conventional machine learning techniques. The proposed system reduces manual effort, improves identification speed, and provides an efficient and user-friendly solution for students, researchers, farmers, and healthcare enthusiasts. Overall, the project highlights the effective application of artificial intelligence and deep learning in botanical research and medicinal plant preservation.
Keywords: Medicinal Plants, Deep Learning, Convolutional Neural Network, Image Classification, Plant Identification, Artificial Intelligence, Firebase Firestore.
Abstract
Adaptive Federated Threat Detection
Khushi D Kunte, Ranganath S R
DOI: 10.17148/IJARCCE.2026.15581
Abstract: Food safety has become a major concern due to the increasing consumption of packaged and processed food products containing harmful additives and chemicals. Consumers often find it difficult to understand ingredient labels and identify the health risks associated with food preservatives, artificial colours, and additives. Conventional food awareness systems provide limited real-time analysis and lack intelligent recommendation mechanisms. This project proposes Caution Bites, an AI-powered food safety and ingredient analysis system that helps users identify harmful food additives and make healthier dietary decisions. The system allows users to scan or enter food ingredients, after which the application analyses the content using intelligent data processing and AI-based risk evaluation. It provides detailed information regarding additives, preservatives, allergens, and potential health impacts. The framework also generates safer food recommendations and health alerts based on detected ingredients. Experimental evaluation demonstrates improved consumer awareness, fast ingredient analysis, and effective health risk identification while maintaining a user- friendly and scalable architecture.
Keywords: Food Safety, AI-Based Ingredient Analysis, Health Risk Detection, Additive Identification, Intelligent Recommendation System, Consumer Health Monitoring
Keywords: Food Safety, AI-Based Ingredient Analysis, Health Risk Detection, Additive Identification, Intelligent Recommendation System, Consumer Health Monitoring
Abstract
Skin Disease Classification Using CNN
Amritha R, Dr. H K Madhu
DOI: 10.17148/IJARCCE.2026.15582
Abstract: Skin diseases represent a significant global healthcare challenge, with certain malignant lesions such as melanoma requiring early diagnosis to improve patient survival rates. Traditional dermatological diagnosis primarily depends on visual examination by specialists, which may be subjective, time-consuming, and limited by the availability of experienced dermatologists. This paper presents a deep learning–based multi-class skin disease classification system using Convolutional Neural Networks (CNNs) for automated analysis of dermoscopic skin images. The proposed framework classifies skin lesions into nine distinct disease categories using a publicly available dermoscopic image dataset. Image preprocessing techniques including resizing, normalization, and augmentation are applied to improve model generalization and classification performance. To enhance interpretability, Explainable Artificial Intelligence (XAI) techniques based on Gradient-weighted Class Activation Mapping (Grad-CAM) are incorporated to visualize important lesion regions influencing prediction outcomes. The trained CNN model is integrated into a Streamlit-based web application that enables real-time image upload and disease prediction along with confidence visualization. Experimental evaluation demonstrates that the proposed system achieves reliable multi-class classification performance with a validation accuracy of approximately 87%. The proposed framework serves as an intelligent decision-support tool for educational and preliminary diagnostic assistance and highlights the potential of explainable deep learning techniques in medical image analysis.This paper discusses the system architecture, methodology, implementation, and performance evaluation of the proposed solution.
Keywords: Skin Disease Classification, Deep Learning, Convolutional Neural Network, Explainable AI, Grad-CAM, Dermoscopic Image Analysis, Streamlit
Keywords: Skin Disease Classification, Deep Learning, Convolutional Neural Network, Explainable AI, Grad-CAM, Dermoscopic Image Analysis, Streamlit
Abstract
Cisix: Technology for Net-Zero Carbon Ecosystem
Darshan D Kolmi, Goutham E, Hrishikesh Bodapatti, Javeed Hussain P, Dr. Anita Patil, Vijay Kumar
DOI: 10.17148/IJARCCE.2026.15583
Abstract: Farmers represent one of the largest untapped sources of carbon sequestration potential globally, yet they remain almost entirely excluded from carbon markets. This paper presents Cisix, a technology-mediated carbon trading platform designed to bridge the structural gap between rural agricultural producers, Sustainable project Developers and Carbon credit buyers. Unlike conventional carbon market infrastructure — which demands technical literacy, legal capacity, and upfront capital that most smallholder farmers simply do not have, Cisix embeds farmer training and certification directly into the trading pipeline, treating knowledge transfer not as a peripheral service but as a core market function. Drawing on platform design principles and inclusive finance frameworks, we examine how Cisix addresses three compounding barriers to farmer participation: awareness deficits around carbon credit mechanisms, absence of accessible verification pathways, and the intermediary-heavy structures that erode farmer earnings. The platform operationalizes a four-stage model — assessment, training, credit generation, and direct marketplace access of ours — supported by multilingual outreach and website which supports at local village areas. Early deployment data suggest that co-locating financial participation with structured capacity building significantly improves both onboarding rates and long-term farmer retention in carbon programs. We argue that carbon market inclusion is not merely a technical challenge but a design and governance problem, and that platforms which treat farmers as informed economic agents — rather than passive land managers — are better positioned to deliver durable climate and livelihood outcomes.
Keywords: Carbon credits, Carbon Trading platform, Voluntary Carbon Market, AgriTech, Farmer Inclusion, Sustainable Projects, Rural Capacity Building, Climate Action, Indian Carbon Trading Platform.
Keywords: Carbon credits, Carbon Trading platform, Voluntary Carbon Market, AgriTech, Farmer Inclusion, Sustainable Projects, Rural Capacity Building, Climate Action, Indian Carbon Trading Platform.
Abstract
Recommendation system for cloud service on Trust management services
Deepak BN, Anusha J, Krishnaveni A, Karthik R
DOI: 10.17148/IJARCCE.2026.15584
Abstract: The trust management is challenging issue in cloud computing now a days. The main obstacle for growth and adoption of cloud computing is trust. Though many works has been proposed, the determination of credibility for trust feedbacks is ignored. In this paper, the Cloud Armor, a reputation and recommendation trust management is proposed in which set of functions are provided to delivery of Trust as a Service. Cloud administration faces significant challenge. The Service Level Agreements (SLAs) are facing difficulty in finding trust between cloud service customers and cloud service providers. Buyers’ input can be sensible to the cloud administration characteristic to survey.
Keywords: Trust Management,RecommendationTrust,ReputationTrust,CloudService,TrustResults.
Keywords: Trust Management,RecommendationTrust,ReputationTrust,CloudService,TrustResults.
Abstract
Intelligent Crop Disease Detection Systems: A Review of Deep Learning Approaches
Siddessh K S, Anita Patil
DOI: 10.17148/IJARCCE.2026.15585
Abstract: Agriculture plays a crucial role in global food production, but crop diseases continue to cause major losses in yield and quality each year. Traditional disease detection methods depend on manual inspection by agricultural experts, which is time-consuming, costly, and often ineffective for large-scale farming. Recent advancements in Artificial Intelligence (AI), Deep Learning, Computer Vision, and Internet of Things (IoT) technologies have enabled the development of intelligent crop disease detection systems capable of identifying plant diseases automatically and accurately. This paper presents a comprehensive review of AI-based crop disease detection approaches using Convolutional Neural Networks (CNN), image processing techniques, mobile applications, and drone-based monitoring systems. The study examines commonly used datasets, preprocessing methods, deep learning architectures, and deployment platforms in modern smart agriculture applications. A four-tier taxonomy is proposed to classify crop disease detection systems based on their level of automation and intelligence. Performance metrics such as accuracy, precision, recall, F1-score, and computational efficiency are also analyzed. Comparative analysis shows that while deep learning models provide high detection accuracy, challenges such as dataset imbalance, varying environmental conditions, internet dependency, and scalability still remain unresolved. Finally, the paper identifies major research gaps and discusses future directions toward intelligent AI-powered precision agriculture systems.
Abstract
Intelligent Fingerprint Storage and Management System for Authentication Applications on Cloud Storage
Vaasthava Sree Sai Reddy N, Subhamay Parya, Spoorti Patil, Vaishnavi Reddy, Dr.SoniaMariaD'souza
DOI: 10.17148/IJARCCE.2026.15586
Abstract: Data security has become highly necessary in our digital era. This paper proposes a Fingerprint Storage System (FSS) on cloud storage that provides users a secure way to access their files through fingerprint authentication and not password-based authentication. The OFR is used to recognize and authenticate users. Files stored are also encrypted via AES-256 algorithm in order to prevent any unauthorized access and data breach. This paper eliminates the threat of password thefts, phishing, and unauthorized access since passwords would no longer be necessary. The proposed solution is intended to operate in practical applications such as banking, health care, educational institutes, and enterprises.
Keywords: Cloud Technology, Fingerprint accessibility, AES-256 encryption, Biometric Security,Secure Cloud Storage
Keywords: Cloud Technology, Fingerprint accessibility, AES-256 encryption, Biometric Security,Secure Cloud Storage
Abstract
A Study on the Architecture and Operation of Intelligent Knowledge Platforms Using LLMs and AIOps
Shilpa M R, Swetha C S
DOI: 10.17148/IJARCCE.2026.15587
Abstract: Traditional knowledge platforms often rely on keyword-based search and manual system management, resulting in poor contextual understanding and operational inefficiencies. This paper presents a study on the architecture and operation of intelligent knowledge platforms integrating Large Language Models (LLMs) and AIOps technologies. LLMs enable semantic understanding, contextual reasoning, and intelligent response generation, while AIOps enhances operational reliability through automated monitoring, anomaly detection, predictive analysis, and self-healing mechanisms. The proposed layered architecture combines semantic intelligence with operational intelligence to improve scalability, user experience, and system resilience. The study also discusses key challenges, applications, and future research directions for AI-driven knowledge platforms.
Keywords: Large Language Models (LLMs), AIOps, Intelligent Knowledge Platforms, Semantic Intelligence, Automated IT Operations, Artificial Intelligence, Context-Aware Systems, Intelligent Automation
Keywords: Large Language Models (LLMs), AIOps, Intelligent Knowledge Platforms, Semantic Intelligence, Automated IT Operations, Artificial Intelligence, Context-Aware Systems, Intelligent Automation
Abstract
AI-Based UPI Fraud Detection Using Machine Learning and Real-Time Analysis
Atul Shivaji Kamble
DOI: 10.17148/IJARCCE.2026.15588
Abstract: The rapid adoption of Unified Payments Interface (UPI) has revolutionized digital transactions in India. However, this growth has also led to an increase in fraudulent activities such as phishing, fake payment requests, and unauthorized transactions. Traditional fraud detection systems are often reactive and fail to prevent fraud in real time.
This paper presents an AI-based UPI fraud detection system that uses machine learning techniques to identify suspicious transactions before they are completed. The system analyzes transaction patterns, detects anomalies, and generates real- time alerts for users. A web-based dashboard provides users with insights into their transaction behavior, while Firebase ensures real-time data synchronization and scalability. The proposed system focuses on proactive fraud prevention, improving user trust and enhancing the security of digital payment systems.
Keywords: UPI, Fraud Detection, Machine Learning, Real-Time Analysis, AI, Firebase.
This paper presents an AI-based UPI fraud detection system that uses machine learning techniques to identify suspicious transactions before they are completed. The system analyzes transaction patterns, detects anomalies, and generates real- time alerts for users. A web-based dashboard provides users with insights into their transaction behavior, while Firebase ensures real-time data synchronization and scalability. The proposed system focuses on proactive fraud prevention, improving user trust and enhancing the security of digital payment systems.
Keywords: UPI, Fraud Detection, Machine Learning, Real-Time Analysis, AI, Firebase.
Abstract
Open Source Software Licenses: A Comparative Analysis of GPL, MIT, and Apache Manjunath S Rakaraddi, Bhairam V Pawar, Rahul M, Raviteja Javali,
Muhibur Rahman T. R
DOI: 10.17148/IJARCCE.2026.15589
Abstract: Open-source software licensing plays a vital role in modern software development by defining the legal permissions, responsibilities, and limitations associated with the use, modifi-cation, and distribution of software. Among the numerous open-source licenses available, the GNU General Public License (GPL), MIT License, and Apache License 2.0 are widely adopted due to their distinct licensing models and practical significance in both academic and industrial environments. This paper presents a comparative study of these three major open-source licenses by analyzing their features, permissions, restrictions, distribution policies, and patent considerations. The GPL license emphasizes software freedom through its copyleft approach, requiring deriva-tive works to remain open source, whereas the MIT License pro-vides maximum flexibility with minimal restrictions. The Apache License 2.0 combines permissive licensing with explicit patent protection, making it suitable for commercial and enterprise applications. The study highlights the advantages, limitations, and real-world applications of each license to help developers, researchers, and organizations select appropriate licensing strate-gies for software projects. The analysis demonstrates that the choice of an open-source license significantly impacts software collaboration, legal compliance, commercial adoption, and long-term project sustainability.
Keywords: Open-source software, GPL, MIT License, Apache License 2.0, software licensing, copyleft, permissive licenses, patent protection, software compliance
Keywords: Open-source software, GPL, MIT License, Apache License 2.0, software licensing, copyleft, permissive licenses, patent protection, software compliance
Abstract
PlacePrep - Smart DSA Learning and Placement Readiness System
Abhishek P. Nachankar, Neha P. Dhuriya, Amisha Jain, Anisha Katkade, Kumkum Laveti, Ishwari Khandwe, Amartya Choubey, Amit Gond, Atharva Dayalwar
DOI: 10.17148/IJARCCE.2026.15590
Abstract: PlacePrep is a full-stack web application developed to enhance campus placement preparation for IT students through an integrated learning ecosystem. The platform combines DSA practice, aptitude training, AI-based mock interviews, resume building, and progress analytics into a single system. It features a live coding environment, ATS- compatible resume generation, company-wise preparation tools, leader boards, streak tracking, and performance dashboards. Developed using the MERN stack (MongoDB, Express.js, React, and Node.js), the system ensures scalability and modularity through RESTful APIs and cloud integration. PlacePrep aims to improve placement readiness, streamline preparation, and increase student success rates.
Keywords: Campus Placement Preparation, MERN Stack, Artificial Intelligence, Data Structures and Algorithms, Mock Interviews, Resume Builder, ATS Compatibility, Performance Analytics, Web Application, Placement Readiness.
Keywords: Campus Placement Preparation, MERN Stack, Artificial Intelligence, Data Structures and Algorithms, Mock Interviews, Resume Builder, ATS Compatibility, Performance Analytics, Web Application, Placement Readiness.
Abstract
Age Detection Using Machine Learning Techniques: A Comprehensive Review
Akshata Pravin Tatar, Nikita Balasaheb Korde, Kirti Dinkar More
DOI: 10.17148/IJARCCE.2026.15591
Abstract: Age estimation from facial images has become an important research area in computer vision due to its applications in surveillance, biometric authentication, healthcare monitoring, human-computer interaction, and demographic analysis. However, accurate age prediction remains challenging because the human aging process is nonlinear, highly individualized, and influenced by various biological and environmental factors. Early age estimation approaches relied on handcrafted feature extraction techniques combined with traditional machine learning algorithms, which showed limited robustness in unconstrained environments. Recent advancements in deep learning, particularly convolutional neural networks (CNNs), have significantly improved prediction accuracy by enabling automatic feature extraction and hierarchical representation learning from facial images. In addition, techniques such as ordinal regression, label distribution learning, transfer learning, and lightweight deep learning models have further enhanced the performance and efficiency of age estimation systems. This paper presents a comprehensive review of traditional, deep learning, hybrid, and emerging age estimation approaches. The study also analyzes commonly used benchmark datasets, evaluation metrics, major challenges, and recent advancements in the field. Furthermore, issues related to dataset bias, domain adaptation, fairness, and ethical concerns are discussed, along with future research directions toward developing reliable, interpretable, and deployable age estimation systems.
Keywords: Age Estimation, Facial Analysis, Machine Learning, Deep Learning, Convolutional Neural Networks, Facial Aging
Keywords: Age Estimation, Facial Analysis, Machine Learning, Deep Learning, Convolutional Neural Networks, Facial Aging
Abstract
AI MEDICAL CHATBOT USING MACHINE LEARNING
Tejaswini KM, Dr. Madhu H.K
DOI: 10.17148/IJARCCE.2026.15592
Abstract: The advancement of Artificial Intelligence (AI) has brought significant changes to the healthcare industry by enabling intelligent systems that can assist users in obtaining medical information efficiently. This project focuses on the design and development of an AI-based Medical Chatbot that provides users with instant medical assistance, preliminary diagnosis, and health-related guidance through an interactive conversational interface. The chatbot is designed to simulate human-like conversation and respond to user queries in natural language, making it accessible and easy to use for individuals without medical or technical knowledge. The AI Medical Chatbot utilizes Natural Language Processing (NLP) techniques to understand user inputs such as symptoms, health concerns, and general medical questions. Based on the analyzed input, the system maps the symptoms to a structured medical knowledge base and generates appropriate responses, including possible health conditions, basic precautions, first-aid suggestions, and lifestyle recommendations. Machine learning algorithms are used to improve the accuracy of responses over time by learning from previous interactions and feedback.
Keywords: Artificial Intelligence (AI) ,Machine Learning ,Medical Chatbot ,Natural Language Processing (NLP) Retrieval-Augmented Generation (RAG) ,Large Language Model (LLM) ,Healthcare Assistant, Vector Database, FAISS, Text Embeddings , Conversational AI.
Keywords: Artificial Intelligence (AI) ,Machine Learning ,Medical Chatbot ,Natural Language Processing (NLP) Retrieval-Augmented Generation (RAG) ,Large Language Model (LLM) ,Healthcare Assistant, Vector Database, FAISS, Text Embeddings , Conversational AI.
Abstract
Defense Framework for Runtime Monitoring and Proactive Data Safeguarding Against Emerging Ransomware Threats
Sarojini P, Reshma G, Mr. M.V. Prabhakaran M.E
DOI: 10.17148/IJARCCE.2026.15593
Abstract: Nowadays, ransomware attacks are increasing rapidly. Ransomware locks or encrypts important files and asks for money to unlock them, leading to financial loss. Many attacks target websites and web applications due to issues like insecure file uploads, weak input validation, and misconfigurations. In the existing system, the theme they mainly employ is a signature-based detection mechanism, which matches files with a pattern of known malware patterns stored in its database. Once again, if the virus is new or its signature does not exist, then it might escape, and an attack can take place. The proposed system offers automatic and real-time defence from ransomware. The system continuously monitors its behaviour instead of relying on known virus patterns. It uses a Gated Recurrent Unit (GRU) neural network to observe system calls, file access activities, and runtime processes to detect both known and unknown ransomware attacks. In this case, the system protects key data in less than a second by employing CTR-Advanced Encryption Standard (AES) encryption with frequently changing keys when any suspicious behaviour is detected but can still be accessed by any legitimate user to ensure data safety. Honeypot file mechanisms are also used for decoying the attacker to attract them and send an alert in case unauthorised access is attempted. The experimental results show that the three-tier security architecture provides strong protection against ransomware attacks, and it continuously analyses the behaviour, monitors and detects unknown threats, and uses honey files to confirm unauthorised and harmful actions performed by a program and prevent data loss. The system works efficiently without slowing down the computer, so it can be used in real time on modern systems.
Keywords: Ransomware, Behavioural Detection, GRU, AES-CTR, Honey files, Web Security, Zero-Day Attacks.
Keywords: Ransomware, Behavioural Detection, GRU, AES-CTR, Honey files, Web Security, Zero-Day Attacks.
Abstract
Lecture Lens: AI Based Student Learning Model
Rutuja Bhute, Prajakta Koravi, Arpita Khot, Sonali Keskar, Samruddhi Jadhav, Prof. S. S. Sangewar
DOI: 10.17148/IJARCCE.2026.15594
Abstract: The students enrolled in modern education systems often find it hard to concentrate in lectures, manage their study material, and comprehend the intricate subject matter being taught to them. These students have to take down lectures manually and engage with the instructor only during the classroom teaching session. This paper suggests an “Lecture Lens: An AI-Based Student Learning Model,” an innovative and intelligent learning system based on the utilization of AI technologies. This proposed learning system uses ML techniques, NLP tools, automatic speech recognition, and automated text summarization to analyze the lectures and provide personalized study resources to the students. The system can be used by the users to convert their lecture audio to text and identify key concepts from the recorded audio files. They can also create concise lecture notes and summary using this intelligent tool and answer questions generated by the system.
Keywords: Artificial Intelligence (AI), Student Learning Model, Natural Language Processing (NLP), Machine Learning (ML), Speech Recognition, Automated Note Generation.
Keywords: Artificial Intelligence (AI), Student Learning Model, Natural Language Processing (NLP), Machine Learning (ML), Speech Recognition, Automated Note Generation.
Abstract
AI Agents as Virtual Software Developers
Rashmitha S, Vidya S
DOI: 10.17148/IJARCCE.2026.15595
Abstract: Artificial Intelligence (AI) agents are emerging as powerful virtual software developers capable of automating various stages of the Software Development Life Cycle (SDLC). Traditional software development processes often require significant human effort in coding, debugging, testing, deployment, and maintenance, leading to increased development time and operational complexity. AI agents use technologies such as Large Language Models (LLMs), machine learning, natural language processing, and autonomous decision-making to assist developers in performing these tasks efficiently. This paper presents a study on the architecture, operation, and applications of AI agents as virtual software developers. The proposed approach highlights how AI agents can improve coding productivity, software quality, collaboration, and automation within modern development environments. The study also discusses the advantages, challenges, limitations, and future scope of AI-driven autonomous software engineering systems.
Keywords: Artificial Intelligence (AI) Agents, Virtual Software Developers, Large Language Models (LLMs), Software Development Life Cycle (SDLC), Autonomous Coding, Intelligent Automation, Machine Learning, AI-Assisted Programming, Software Engineering, Automated Testing and Debugging
Keywords: Artificial Intelligence (AI) Agents, Virtual Software Developers, Large Language Models (LLMs), Software Development Life Cycle (SDLC), Autonomous Coding, Intelligent Automation, Machine Learning, AI-Assisted Programming, Software Engineering, Automated Testing and Debugging
Abstract
A Study on the Architecture and Operation of AI-Based Digital Well-Being Monitoring Systems
Rakshitha P, Swetha C S
DOI: 10.17148/IJARCCE.2026.15596
Abstract: Digital technologies such as smartphones, social media platforms, wearable devices, and online applications have become an essential part of modern life. Although these technologies improve communication, productivity, education, and entertainment, excessive and uncontrolled digital usage has also created several challenges related to mental health, stress, anxiety, sleep disorders, digital addiction, and reduced emotional well-being. Traditional digital well-being systems mainly provide basic monitoring features such as screen time tracking and app usage statistics, which often lack intelligent behavioral analysis, real-time monitoring, and personalized recommendations. This paper presents a study on AI-Based Digital Well-Being Monitoring Systems that utilize Artificial Intelligence, Machine Learning, Deep Learning, wearable sensing technologies, and behavioral analytics to monitor user digital activities and improve overall well-being. The proposed system analyzes behavioral patterns, emotional conditions, physiological signals, smartphone usage habits, and social media interactions to detect unhealthy digital behavior and provide adaptive wellness recommendations.
Keywords: Artificial Intelligence (AI), Digital Well-Being, Machine Learning, Deep Learning, Behavioral Analytics, Mental Health Monitoring, Wearable Devices, Emotion Recognition, Stress Detection, Personalized Recommendation Systems.
Keywords: Artificial Intelligence (AI), Digital Well-Being, Machine Learning, Deep Learning, Behavioral Analytics, Mental Health Monitoring, Wearable Devices, Emotion Recognition, Stress Detection, Personalized Recommendation Systems.
Abstract
AgroSense II: Smart Plant Disease Detection and Treatment Recommender
Hm Mujahid Pasha, B Prem Kumar, Md Mohseen, Rohit M, Dr. Anita Patil, Mr. Pavan Kumar
DOI: 10.17148/IJARCCE.2026.15597
Abstract: Plant diseases are one of the major causes of crop loss and reduced agricultural productivity worldwide. Farmers, especially in rural regions, often struggle to identify diseases at an early stage due to limited access to expert support and modern agricultural tools. Traditional diagnosis methods depend on manual observation, which is time- consuming and prone to errors in large-scale farming. This paper presents AgroSense-II: AI-Powered Plant Disease Detection and Treatment Recommender, a unified web-based platform that integrates machine learning-based disease prediction, treatment recommendation, analytics dashboards, and historical record management into a single lightweight framework. The system employs a Random Forest classifier trained on agricultural datasets, a FastAPI-powered backend, ReactJS frontend, and MongoDB for data storage. By combining these technologies into one deployable architecture, AgroSense-II helps farmers identify diseases faster, receive actionable treatment guidance, and make smarter crop management decisions without depending on expensive hardware infrastructure.
Keywords: Plant Disease Detection, Precision Agriculture, Machine Learning, Random Forest Classifier, Smart Farming, FastAPI, ReactJS, MongoDB, Crop Monitoring, Treatment Recommendation.
Keywords: Plant Disease Detection, Precision Agriculture, Machine Learning, Random Forest Classifier, Smart Farming, FastAPI, ReactJS, MongoDB, Crop Monitoring, Treatment Recommendation.
Abstract
AI Powered Drowning Detection and Alert System for Swimming Pool
Phanindra Reddy K, Akshatha T, B Akshitha, Gouri, G Poojitha
DOI: 10.17148/IJARCCE.2026.15598
Abstract: The use of Artificial Intelligence in intelligent surveillance systems has improved the ability to detect emergency situations automatically and efficiently. This project presents an AI Powered Human Drowning Detection and Alert System designed for swimming pool safety using Computer Vision and Deep Learning techniques. The proposed system continuously monitors uploaded media files and live webcam streams to identify possible drowning situations in real time.
The system is implemented using a custom-trained YOLOv8 model integrated with OpenCV for image and video processing. Flask is used for backend processing, while HTML, JavaScript, and Tailwind CSS are used to develop the frontend monitoring interface. The system performs frame-by-frame analysis to detect swimmers, generate bounding boxes, and classify activities based on prediction confidence.
To improve detection reliability, the system applies confidence thresholding, frame buffering, and danger percentage analysis to reduce false alerts caused by temporary movements or water disturbances. Whenever dangerous activity is detected continuously, the system immediately generates warning notifications and audio alerts.
The developed solution provides a practical, affordable, and real-time safety monitoring system for swimming pools and other aquatic environments. By combining deep learning, intelligent surveillance, and automated alert generation, the system helps improve swimmer safety and reduces emergency response time.
Keywords: Artificial Intelligence, YOLOv8, Drowning Detection, Computer Vision, OpenCV, Deep Learning, Real- Time Monitoring, Surveillance System, Swimming Pool Safety, Alert System.
The system is implemented using a custom-trained YOLOv8 model integrated with OpenCV for image and video processing. Flask is used for backend processing, while HTML, JavaScript, and Tailwind CSS are used to develop the frontend monitoring interface. The system performs frame-by-frame analysis to detect swimmers, generate bounding boxes, and classify activities based on prediction confidence.
To improve detection reliability, the system applies confidence thresholding, frame buffering, and danger percentage analysis to reduce false alerts caused by temporary movements or water disturbances. Whenever dangerous activity is detected continuously, the system immediately generates warning notifications and audio alerts.
The developed solution provides a practical, affordable, and real-time safety monitoring system for swimming pools and other aquatic environments. By combining deep learning, intelligent surveillance, and automated alert generation, the system helps improve swimmer safety and reduces emergency response time.
Keywords: Artificial Intelligence, YOLOv8, Drowning Detection, Computer Vision, OpenCV, Deep Learning, Real- Time Monitoring, Surveillance System, Swimming Pool Safety, Alert System.
Abstract
AI-Foresight: Market Research and Trend Analyzer
Akshata Panchal, Vidya S
DOI: 10.17148/IJARCCE.2026.15599
Abstract: In the modern digital economy, organizations generate and consume massive volumes of unstructured textual data through online news portals, blogs, forums, and social media platforms. Extracting meaningful insights from such data using traditional market research techniques is time-consuming, expensive, and often inefficient. This paper presents AI-Foresight – Market Research and Trend Analyzer, an AI-driven platform developed to automate market intelligence generation using Natural Language Processing (NLP), Machine Learning (ML), and Generative AI techniques. The proposed system performs automated data collection, text preprocessing, sentiment analysis, topic modelling, trend detection, and AI-based summarization to transform raw textual information into actionable business insights. The platform integrates tools such as BeautifulSoup, NLTK, spaCy, BERTopic, VADER, TextBlob, and GPT-based summarization models, while presenting analytical outputs through an interactive Streamlit dashboard with dynamic visualizations. Experimental evaluation demonstrates that the system efficiently processes large-scale textual datasets, accurately identifies market trends, and significantly reduces manual analytical effort. The proposed solution provides organizations, researchers, and business strategists with a scalable and intelligent framework for real-time market analysis and strategic decision-making.
Keywords: Artificial Intelligence, Market Research, Trend Analysis, Natural Language Processing, Sentiment Analysis, Topic Modelling, Generative AI, Business Intelligence.
Keywords: Artificial Intelligence, Market Research, Trend Analysis, Natural Language Processing, Sentiment Analysis, Topic Modelling, Generative AI, Business Intelligence.
Abstract
Department of Computer Science and Engineering, Arasu Engineering College Prevention and Detection of Botnet Attacks using Double layered machine learning Technique
S. Parvathy, S. Mounika, M. Nihidha, M. Sruthi
DOI: 10.17148/IJARCCE
Abstract: In multi level botnet attack in prevailing cyber attacks in the IoT environment starts and ends detection activities. In existing detection of botnet attacks compromising the IoT devices initially performs ddos attacks. According to the various performances of existing machine learning botnet detection model is limited to the trained data which are already specified. The consequences towards the datasets according to the diversified attack patterns that performs perfectly will be questionable. In our proposed methodology the generalized scanning of datasets in DDoS attacks generates 33 varieties of detection patterns. Integration of detecting samples of DDoS at tacks with publicly available datasets within the limit of more attacks. Proposed prevention and detection of double layered machine learning techniques helps in training the dataset models. Prior to the attacking stage the IoT botnet attacks identifies from the trained double layered attack identification and detection models. In the next layer efficiency of datasets detection approach with more accuracy and precise training models will be provided.
Keywords: Botnet Attack, Cyber Attacks, Datasets, Double Layered Machine Learning Techniques, Training Datasets.
Keywords: Botnet Attack, Cyber Attacks, Datasets, Double Layered Machine Learning Techniques, Training Datasets.
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
Animal Tracking & Detection System
Sushilkumar Dasharath Birunagi, Pranav Pandurang Bhosale, Prathmesh Sukhadev Patil, Rushikesh Satish Mulik, Prof. A.B. Jagadale
DOI: 10.17148/IJARCCE.2026.155100
Abstract: The Animal Tracking and Detection System is an advanced smart solution designed to improve farm management by automatically monitoring animal movement and detecting their presence in real time. It helps farmers track livestock location, prevent crop damage, and enhance farm security using IoT-based technologies. The system helps farmers locate livestock, avoid losses, and protect crops through continuous monitoring and instant notifications. It minimizes manual effort, improves monitoring accuracy, and enhances overall farm management.
Keywords: Animal Tracking & Detection, GPS Module, ESP32, Livestock Monitoring, Farm Security, Real-Time Location, MIT App Inventor, Smart Agriculture.
Keywords: Animal Tracking & Detection, GPS Module, ESP32, Livestock Monitoring, Farm Security, Real-Time Location, MIT App Inventor, Smart Agriculture.
Abstract
AI Powered Phishing Email Detector with Gmail Live Scanning
Mr. M. V. Prabhakaran, Mugil M U, Srikanth T
DOI: 10.17148/IJARCCE.2026.155101
Abstract: Phishing attacks remain the most prevalent cyber threat, responsible for over 90% of data breaches and causing billions of dollars in financial losses annually. Traditional rule- based email filters fail to detect sophisticated, AI- generated phishing emails that evade signature-based detection. This paper presents PhishGuard, a novel hybrid dual- stage classification framework that combines Random Forest ensemble learning with a CNN-LSTM deep neural network for real-time phishing email detection. Our system extracts 25 engineered features from email headers and content, including SPF/DKIM authentication status, URL analysis, and linguistic patterns. The framework employs a weighted decision fusion mechanism that adjusts confidence scores based on sender authentication and trusted domain verification. We implement a complete web-based solution with FastAPI backend and React frontend, featuring seamless Gmail API integration via OAuth 2.0 for real-time inbox scanning. Experimental results demonstrate that PhishGuard achieves 96.3% accuracy, 96.1% precision, and 96.5% recall on our evaluation dataset, outperforming single-model approaches by 2.1%. The system processes emails in under 165ms, making it suitable for real-time deployment. Our contribution includes the complete open-source implementation, a comprehensive feature engineering pipeline, and a user-friendly interface that educates users about phishing indicators while protecting them.
Keywords: Phishing Detection, Deep Learning, Random Forest, CNN-LSTM, Email Security, Natural Language Process- ing, Gmail API, Cybersecurity
Keywords: Phishing Detection, Deep Learning, Random Forest, CNN-LSTM, Email Security, Natural Language Process- ing, Gmail API, Cybersecurity
Abstract
Detection of Mobile Malware (Android) using Machine Learning and Hybrid Analysis
Mrs. Uma S, Rahini R, Rudhramitra S
DOI: 10.17148/IJARCCE.2026.155102
Abstract: The exponential proliferation of mobile devices has catalyzed a parallel surge in Android malware, presenting critical security challenges in information technology. With Android commanding the lion’s share of the global mobile operating system market, it has become the primary target for malicious actors employing sophisticated evasion techniques such as dynamic code loading, reflection, and automated repackaging (obfuscation). Detecting zero-day malware—attacks that exploit previously unknown vulnerabilities—has thus become a paramount objective for security researchers. Traditional detection paradigms, which predominantly leverage signature-based static analysis, are increasingly rendered ineffective by polymorphic malware. Conversely, dynamic analysis, while robust, incurs prohibitive computational overhead, rendering it unsuitable for real-time, on-device application.
In this paper, we propose a novel, intelligent, two-stage hybrid framework that synergizes the efficiency of Deep Learning with the forensic depth of Hybrid Analysis. The proposed model operates on a ”Filter and Focus” principle. The first stage acts as a high-speed filter, employing a 1D Convolutional Neural Network (CNN) to analyze vectorized API call graphs extracted via FlowDroid. To address the ”black box” nature of neural networks, we integrate Gradient-weighted Class Activation Mapping (Grad-CAM) to provide visual explainability of malicious triggers. Furthermore, a Jaccard Similarity module compares these features against known threat signatures. Only applications classified as ’Benign’ or ’Uncertain’ by this stage are forwarded to the second stage, which employs a rigorous hybrid engine combining Mobile Security Framework (MobSF) for static deepinspection and Quark-Engine for dynamic behavioral graphing. Experimental results on a diverse dataset of 13,298 applications (including obfuscated samples from PRAGuard) demonstrate that our hybrid model achieves an accuracy of 97.79%, significantly outperforming standalone Deep Belief Networks (DBN) and Gated Recurrent Units (GRU). The system drastically reduces false positives while maintaining a low average latency, making it a viable solution for scalable, real-world Android security.
Keywords: Android Malware Detection, Network Security, Machine Learning, Static Analysis, Dynamic Analysis, CNN, Grad-CAM, Hybrid Analysis, Deep Belief Network (DBN), Gated Recurrent Unit (GRU), Cyber Threat Intelligence, Opcode Analysis, Adversarial Learning.
In this paper, we propose a novel, intelligent, two-stage hybrid framework that synergizes the efficiency of Deep Learning with the forensic depth of Hybrid Analysis. The proposed model operates on a ”Filter and Focus” principle. The first stage acts as a high-speed filter, employing a 1D Convolutional Neural Network (CNN) to analyze vectorized API call graphs extracted via FlowDroid. To address the ”black box” nature of neural networks, we integrate Gradient-weighted Class Activation Mapping (Grad-CAM) to provide visual explainability of malicious triggers. Furthermore, a Jaccard Similarity module compares these features against known threat signatures. Only applications classified as ’Benign’ or ’Uncertain’ by this stage are forwarded to the second stage, which employs a rigorous hybrid engine combining Mobile Security Framework (MobSF) for static deepinspection and Quark-Engine for dynamic behavioral graphing. Experimental results on a diverse dataset of 13,298 applications (including obfuscated samples from PRAGuard) demonstrate that our hybrid model achieves an accuracy of 97.79%, significantly outperforming standalone Deep Belief Networks (DBN) and Gated Recurrent Units (GRU). The system drastically reduces false positives while maintaining a low average latency, making it a viable solution for scalable, real-world Android security.
Keywords: Android Malware Detection, Network Security, Machine Learning, Static Analysis, Dynamic Analysis, CNN, Grad-CAM, Hybrid Analysis, Deep Belief Network (DBN), Gated Recurrent Unit (GRU), Cyber Threat Intelligence, Opcode Analysis, Adversarial Learning.
Abstract
Insider Threat Detection System Using Machine Learning
Mr. M.V. Prabhakaran, Naveen A, Saran R, HOD, Computer Science Engineering (Cyber Security)
DOI: 10.17148/IJARCCE.2026.155103
Abstract: Insider threats are one of the major challenges in securing an organizational network since the insider possesses proper access authority, making it hard to trace their malicious activities using traditional security measures. Currently, most insider threat detection systems are based on supervised learning methodologies, which demand a lot of labeled data, most of which tends to be imbalanced. To tackle these problems, this research work will employ a hybrid insider threat detection model that combines Isolation Forest with temporal behavior profiling and with random forest algorithm for classification. The proposed solution is based on simulating normal user activity and identifying irregularities, which could be associated with insider attacks. As opposed to conventional solutions, which rely solely on static variables, this solution uses temporal behavioral variables, including access rate, session time, abnormal system activity during offline sessions, and sudden changes in user activity patterns. The Isolation Forest algorithm is leveraged to identify abnormal activity at the algorithmic level without relying on resampling, thus mitigating overfitting issues and distorting the data. In this paper, the proposed approach has been tested using the CERT insider threat dataset, known to be an extreme class imbalance. Results show that the hybrid approach of the model, which combines the benefits of temporal profiling with the strengths of anomaly-based approaches, greatly increases the accuracy level while at the same time ensuring low false positives. This indicates that the model can be said to be quite robust.
Keywords: Insider threat detection, anomaly detection, Isolation Forest, Random forest, behavioral analysis, temporal profiling, imbalanced dataset, framework.
Keywords: Insider threat detection, anomaly detection, Isolation Forest, Random forest, behavioral analysis, temporal profiling, imbalanced dataset, framework.
Abstract
AI-Based First Aid Chatbot Using TF-IDF and Maximum Marginal Relevance for Efficient Symptom-Based Assistance
Vinit Sangoi, Sachi Pandya, Het Shah, Harsh Shinde, Dr. Manimala Mahato
DOI: 10.17148/IJARCCE.2026.155104
Abstract: During medical emergencies, having quick access to first aid information is essential, particularly if professional assistance is not readily available. This paper describes a First Aid Chatbot that uses user-reported symptoms to provide basic medical advice. Through a web-based chatbot interface, the system enables users to enter symptoms and provides suitable first aid recommendations that are divided into two categories: immediate remedies and whole-day care recommendations. The chatbot uses methods like Maximum Marginal Relevance (MMR) and deduplication to eliminate repetitive suggestions while processing user inputs and retrieving pertinent responses from a predefined remedy dataset. The system, which is built with Node.js, Express.js, MongoDB, and EJS, attempts to increase accessibility to fundamental medical advice and raise awareness of first aid procedures.
Keywords: First Aid Chatbot, Healthcare Chatbots, Symptom Analysis, Conversational AI,
Keywords: First Aid Chatbot, Healthcare Chatbots, Symptom Analysis, Conversational AI,
Abstract
Detection of Blackhole and Sinkhole Attacks in Wireless Sensor Networks Using a Lightweight Secure Protocol
C. Karthika and Dr. P. E. Irin Dorathy
DOI: 10.17148/IJARCCE.2026.155105
Abstract: Wireless Sensor Networks (WSNs) are widely used in critical applications such as environmental monitoring, healthcare, and industrial automation, where secure and reliable data transmission is essential. However, due to resource constraints and unattended deployment, WSNs are highly vulnerable to routing attacks such as blackhole and sinkhole attacks. This paper proposes a simple and lightweight trust-based security protocol designed to detect and isolate malicious nodes with minimal computational and communication overhead. The protocol operates in three key stages: neighbour monitoring, trust evaluation, and secure route selection. In the monitoring phase, nodes locally observe the packet forwarding behaviour of their one-hop neighbours. A combined trust score is then computed using forwarding reliability and traffic consistency metrics to accurately identify malicious behaviour. Nodes with low trust scores are isolated through a distributed blacklist mechanism. Finally, secure routing decisions are made by selecting nodes with high trust values and sufficient residual energy, ensuring both reliability and energy efficiency. The proposed approach effectively detects both blackhole and sinkhole attacks while maintaining low overhead, making it suitable for resource- constrained WSN environments.
Keywords: WSN, Secure Routing, Lightweight Protocol, Forwarding Ratio, Malicious Node Detection.
Keywords: WSN, Secure Routing, Lightweight Protocol, Forwarding Ratio, Malicious Node Detection.
Abstract
POWER OF SDK
Ashwini P N, Dr. H K Madhu
DOI: 10.17148/IJARCCE.2026.155106
Abstract: In the modern mobile ecosystem, understanding user behavior is critical for application growth and monetization. Software Development Kits (SDKs) serve as the fundamental bridge between raw application interactions and actionable marketing insights. This project explores the architectural implementation and strategic impact of data- tracking SDKs in mobile applications. By integrating specialized tracking modules, developers can capture granular user actions—ranging from button clicks to session duration—without building complex backend infrastructure from scratch.
The proposed framework demonstrates how SDKs utilize event-driven architectures to gather behavioral data, which is then processed to create detailed user personas. We examine the role of SDKs in audience segmentation, enabling marketers to target specific demographics through automated triggers and personalized push notifications. Key features such as event logging, user profiling, and real-time analytics are analyzed for their efficacy in increasing conversion rates. Furthermore, this study addresses the balance between deep data harvesting and user privacy compliance (GDPR/CCPA). The results highlight that well-implemented SDKs reduce development overhead by 40% while significantly enhancing the precision of marketing campaigns through data-driven audience targeting.
Keywords: Software Development Kit (SDK), Data Tracking, User Behavior, Marketing Analytics, Audience Targeting, Event-Driven Architecture, Mobile Marketing
The proposed framework demonstrates how SDKs utilize event-driven architectures to gather behavioral data, which is then processed to create detailed user personas. We examine the role of SDKs in audience segmentation, enabling marketers to target specific demographics through automated triggers and personalized push notifications. Key features such as event logging, user profiling, and real-time analytics are analyzed for their efficacy in increasing conversion rates. Furthermore, this study addresses the balance between deep data harvesting and user privacy compliance (GDPR/CCPA). The results highlight that well-implemented SDKs reduce development overhead by 40% while significantly enhancing the precision of marketing campaigns through data-driven audience targeting.
Keywords: Software Development Kit (SDK), Data Tracking, User Behavior, Marketing Analytics, Audience Targeting, Event-Driven Architecture, Mobile Marketing
Abstract
Wallify – Make walls talk, make vibes pop.
Dr. Pravin Shinde, Saniya Shah, Vishvesh Ghadge, Moksh Gala, Akshi Gori
DOI: 10.17148/IJARCCE.2026.155107
Abstract: Interior wall decoration plays a crucial role in defining the aesthetic identity and emotional ambiance of indoor spaces. However, users often struggle not with visualization, but with ideation—deciding what to decorate, how to style it, and which combinations best suit their preferences and environment. This paper presents Wallify, an Artificial Intelligence- based interior decor ideation platform that assists users in generating personalized wall decor concepts based on contextual inputs such as user intent, style preferences, spatial characteristics, and mood. Instead of focusing primarily on augmented visualization, the system emphasizes intelligent recommendation and design reasoning by transforming user inputs into structured decor concepts, categorized themes, and curated layout suggestions. By integrating AI-driven recommendation models, design heuristics, and aesthetic analysis, Wallify acts as a digital decor advisor that simplifies decision-making, enhances creativity, and enables users to conceptualize cohesive interior designs efficiently.
Keywords: Augmented Reality, 3D Parallax Visualization, Interior Decoration, Event Decor, Artificial Intelligence, Computer Vision, Wall Detection, Image Processing, Personalized Recommendation System, Virtual Visualization, Smart Design Systems, True-Scale Calibration.
Keywords: Augmented Reality, 3D Parallax Visualization, Interior Decoration, Event Decor, Artificial Intelligence, Computer Vision, Wall Detection, Image Processing, Personalized Recommendation System, Virtual Visualization, Smart Design Systems, True-Scale Calibration.
Abstract
MCP: Multi-Tenant Chat Support System with AI Integration
Preksha Dewoolkar, Chirag Patankar, Samruddhi Pande, Dr. Rahul Pachade
DOI: 10.17148/IJARCCE.2026.155108
Abstract: Language models have proved their capability of comprehending human utterances. If such models are used for assisting clients in businesses, this poses another challenge for the model user. The challenge arises from the fact that large language models make claims that are attractive and seem correct, but actually are not. This phenomenon is referred to as "hallucination.".
To solve this problem we use something called Retrieval-Augmented Generation. This is how it works: the model looks at information from an area before it answers a question about that information. We made a system called MCP that uses Retrieval-Augmented Generation. MCP is a platform that businesses can use to talk to their customers. The main thing about MCP is that it keeps each businesss information from the other businesses. MCP can also handle a lot of documents, for each business. It even keeps track of how each business is using the MCP system. This way the businesses can see how often they are using MCP to talk to their customers. We tried out MCP. It worked well. The model was much better, at giving answers and did not make things up as much as other models do. This is because it was looking at information before answering questions.
Keywords: Retrieval-Augmented Generation, Multi-Tenant Architecture, Large Language Models, Vector Databases, Customer Support Systems, Semantic Search.
To solve this problem we use something called Retrieval-Augmented Generation. This is how it works: the model looks at information from an area before it answers a question about that information. We made a system called MCP that uses Retrieval-Augmented Generation. MCP is a platform that businesses can use to talk to their customers. The main thing about MCP is that it keeps each businesss information from the other businesses. MCP can also handle a lot of documents, for each business. It even keeps track of how each business is using the MCP system. This way the businesses can see how often they are using MCP to talk to their customers. We tried out MCP. It worked well. The model was much better, at giving answers and did not make things up as much as other models do. This is because it was looking at information before answering questions.
Keywords: Retrieval-Augmented Generation, Multi-Tenant Architecture, Large Language Models, Vector Databases, Customer Support Systems, Semantic Search.
Abstract
VR MEDICAL TRAINING: AN IMMERSIVE AND INTERACTIVE PLATFORM FOR SURGICAL SIMULATION AND ANATOMICAL EXPLORATION
Cinita Mary Mathew, George Mathew, Alen James, Rehan Siby Joseph
DOI: 10.17148/IJARCCE.2026.155109
Abstract: VR Medical Training offers a realistic, risk-free environment where trainees can practice and refine their skills, including surgical techniques and procedural tasks. It is highly useful as a visualization aid, helping trainees understand complex medical procedures through interactive learning. The developed system provides a real-time 3D representation of various body parts in an immersive VR environment, offering structured learning tools through multiple interfaces and functionalities. Unity 3D serves as the foundation for creating the interactive experience, while Blender is utilized to develop anatomically accurate 3D models and animations of body parts and medical instruments. The key features of the application include instructions and tutorials displayed on a canvas, along with an incision guide for precise procedural practice. The platform significantly enhances medical training by providing an engaging, interactive, and effective learning experience.
Keywords: VR Training, Immersive Environment, Interactive Learning, Surgical Simulation
Keywords: VR Training, Immersive Environment, Interactive Learning, Surgical Simulation
Abstract
CONCEPTS OF NETWORKING
Mukesh Shantaram Khodke, Mayur Goyekar, Asst. Prof. Namrata Bachhav
DOI: 10.17148/IJARCCE.2026.155110
Abstract: This research focuses on modern enterprise networking technologies and the transition from traditional MPLS- based networks to SD-WAN and cloud-based networking systems. The study covers routing protocols such as OSPF, BGP, and RIP, routing ring topology, network security, and AI-based automation. Network simulations and testing were performed using tools like Cisco Packet Tracer, GNS3, EVE-NG, and Wireshark to analyze latency, packet loss, failover time, and routing performance. The results show that SD-WAN and modern networking technologies provide better scalability, security, faster convergence, improved bandwidth utilization, and enhanced cloud connectivity compared to traditional networks. The research also highlights the growing role of AI and automation in future enterprise networking.
Keywords: SD-WAN, MPLS, OSPF, BGP, RIP, Enterprise Networking, Network Security, Cloud Networking, AI Automation, Routing Protocols, VLAN, Failover, Network Scalability.
Keywords: SD-WAN, MPLS, OSPF, BGP, RIP, Enterprise Networking, Network Security, Cloud Networking, AI Automation, Routing Protocols, VLAN, Failover, Network Scalability.
Abstract
AI-Based Mock Interview System Using Natural Language Processing and Real-Time Feedback
Tushar Patel, Krisha Bhanushali, Drashti Gajara, Samiksha Thakur, Rahul Pachade
DOI: 10.17148/IJARCCE.2026.155111
Abstract: We have created a mock interview tool that attempts to address one particular problem: most interview practice tools pose the same questions to all candidates, but this is not how actual interviews work. Our tool analyzes a candidate’s resume, identifies structured data within it via a DeBERTa model, and then generates questions based on this data via a locally installed Mistral instance via Ollama, with voice interaction via WebRTC and Whisper. After each response, we evaluate five criteria: whether it was relevant, technically accurate, insightful, well-expressed, and confidently stated. Our resume analysis achieved 91% precision and recall, and average response latency was close to 320 ms, but we are naturally a little nervous about how much we should read into these metrics, even from a limited test pool.
Keywords: Artificial Intelligence, Large Language Models, Named Entity Recognition, Natural Language Processing, Voice Activity Detection, WebRTC
Keywords: Artificial Intelligence, Large Language Models, Named Entity Recognition, Natural Language Processing, Voice Activity Detection, WebRTC
Abstract
AQUAVISION:AI POWERED FISH SPECIES DETECTION AND DISEASE ANALYSIS
Ashrith L S, Vidya S
DOI: 10.17148/IJARCCE.2026.155112
Abstract: Misclassification of fish species presents a critical challenge to aquaculture, impacting operational efficiency, biodiversity conservation, and sustainable fisheries management. Accurately identifying fish species is essential for preserving ecological balance and enhancing breeding programs. Similarly, delayed detection of fish diseases can lead to devastating consequences, including mass fish deaths, substantial financial losses, and heightened risks of disease outbreaks in aquaculture farms. Traditional methods for identifying fish species and diagnosing diseases largely depend on manual observation, which can be slow, prone to errors, and impractical for large-scale operations. To address these limitations, our system employs an advanced, automated approach to streamline fish species classification and early disease detection. By leveraging cutting-edge image processing and intelligent pattern recognition techniques, the system delivers fast and accurate results. Enhancements in image quality facilitate superior feature extraction, which improves classification and detection outcomes. This automation minimizes reliance on manual inspections, reducing the likelihood of human errors while enabling aquaculture farmers to monitor fish health more effectively. By integrating intelligent technology into aquaculture management, this solution aims to transform fish farming practices. It supports sustainability by promoting healthier aquatic ecosystems, reducing disease-related losses, and contributing to more efficient and eco- friendly fish farming operations.
Keywords: Artificial Intelligence (AI), Deep Learning, Aquaculture, Fish Species Classification, Fish Disease Detection, Convolutional Neural Network (CNN), Transfer Learning, Computer Vision, Image Processing, EfficientNet, MobileNetV2, Grad-CAM, Explainable AI (XAI), Fish Health Monitoring, Smart Aquaculture, Disease Analysis, Machine Learning, Image Classification, Aquaculture Management, Automated Detection, Flask Web Application, SQLite Database, Real-Time Monitoring, Fish Farming, Aquatic Disease Detection.
Keywords: Artificial Intelligence (AI), Deep Learning, Aquaculture, Fish Species Classification, Fish Disease Detection, Convolutional Neural Network (CNN), Transfer Learning, Computer Vision, Image Processing, EfficientNet, MobileNetV2, Grad-CAM, Explainable AI (XAI), Fish Health Monitoring, Smart Aquaculture, Disease Analysis, Machine Learning, Image Classification, Aquaculture Management, Automated Detection, Flask Web Application, SQLite Database, Real-Time Monitoring, Fish Farming, Aquatic Disease Detection.
Abstract
Smart Mirror Using Raspberry Pi
Prof. R. M. Sahu, Patil Vishal Shrikant, Parde Aditya Babasaheb, Pawar Vaishnavi
DOI: 10.17148/IJARCCE.2026.155113
Abstract: Today we can see that everything around us is becoming smart — smartphones, smartwatches, smart TVs are all part of our daily life. But if we think about the mirror which we have been using since many years, it is still the same traditional mirror used only to see our reflection. So we thought, why not make this traditional mirror smart?
In this project we have developed a Smart Mirror using Raspberry Pi which looks like a normal mirror from outside but displays useful real-time information on its surface. The mirror shows information like weather updates, news feeds, calendar, and current time. We have also integrated Amazon Alexa voice assistant so that the user can interact with the mirror hands-free using voice commands.
One important feature of our smart mirror is the PIR motion sensor. When a person stands in front of the mirror, the display automatically turns on, and when no one is present, it acts as a simple mirror. This also helps in saving power consumption.
The system is built using Raspberry Pi as the main controller, a two-way mirror, and a display monitor. This project shows how a simple household object like a mirror can be made smarter and more useful with the help of IoT and embedded systems at a very low cost.
Keywords: Raspberry Pi, PIR sensor, Alexa Voice Assistant, display/monitor, Real-Time Information, Voice Commands, IoT embedded.
In this project we have developed a Smart Mirror using Raspberry Pi which looks like a normal mirror from outside but displays useful real-time information on its surface. The mirror shows information like weather updates, news feeds, calendar, and current time. We have also integrated Amazon Alexa voice assistant so that the user can interact with the mirror hands-free using voice commands.
One important feature of our smart mirror is the PIR motion sensor. When a person stands in front of the mirror, the display automatically turns on, and when no one is present, it acts as a simple mirror. This also helps in saving power consumption.
The system is built using Raspberry Pi as the main controller, a two-way mirror, and a display monitor. This project shows how a simple household object like a mirror can be made smarter and more useful with the help of IoT and embedded systems at a very low cost.
Keywords: Raspberry Pi, PIR sensor, Alexa Voice Assistant, display/monitor, Real-Time Information, Voice Commands, IoT embedded.
Abstract
A Study on the Data Science Life Cycle and Its Applications in Modern Intelligent Systems
Samarth, Theerthashree G S
DOI: 10.17148/IJARCCE.2026.155114
Abstract: Data Science has become one of the most important technologies in modern computing and intelligent decision-making systems. Organizations generate massive amounts of structured and unstructured data every day, creating the need for efficient techniques to collect, process, analyze, and extract meaningful insights from data. The Data Science Life Cycle provides a systematic framework for solving real-world problems using data-driven approaches. This paper presents a study on the phases of the Data Science Life Cycle, including data collection, data preprocessing, exploratory data analysis, feature engineering, model building, evaluation, deployment, and monitoring. The study explains how these stages work together to transform raw data into actionable insights and intelligent predictions. The paper also discusses applications, advantages, challenges, and future trends in Data Science.
Keywords: Data Science, Machine Learning, Data Analysis, Data Preprocessing, Predictive Modeling, Big Data, Artificial Intelligence, Data Visualization
Keywords: Data Science, Machine Learning, Data Analysis, Data Preprocessing, Predictive Modeling, Big Data, Artificial Intelligence, Data Visualization
Abstract
AI-Based Driver Fatigue Detection and Alert System: A Comprehensive Review
Ashith Shankar Avula, Dr. Muhibur Rahman T R*
DOI: 10.17148/IJARCCE.2026.155115
Abstract: Road accidents caused by driver drowsiness and fatigue represent one of the most critical and preventable safety challenges in modern transportation. The World Health Organization estimates that drowsy driving contributes to over 20% of fatal road crashes globally, imposing enormous human and economic costs. Fatigue impairs cognitive functions such as reaction time, hazard perception, and decision-making in ways that are physiologically comparable to alcohol intoxication, yet far harder for the driver to self-detect. This paper presents a comprehensive review of AI-Based Driver Fatigue Detection and Alert Systems — systems that use computer vision, machine learning, and deep learning to monitor driver facial behaviour in real time and issue timely warnings before cognitive impairment leads to catastrophic outcomes. The proposed system architecture employs a standard webcam to capture live video, applies facial landmark detection using Dlib's 68-point predictor, and computes the Eye Aspect Ratio (EAR) and Mouth Aspect Ratio (MAR) to quantify eyelid closure and yawning frequency. When EAR remains below a predefined threshold for a sustained duration, or when yawn frequency exceeds a critical count within a rolling time window, the system triggers an auditory alert via a buzzer and displays an on-screen warning message. The implementation stack — Python, OpenCV, Dlib, imutils, and pygame — is entirely open-source, cost-effective, and capable of running at real-time frame rates on commodity CPU hardware without requiring specialised GPU acceleration or proprietary embedded systems. This survey further proposes a structured four-tier taxonomy classifying existing fatigue detection architectures by functional sophistication, conducts a curated review of fifteen representative peer-reviewed studies from 2016 to 2024, presents cross-paper comparative analysis across six performance dimensions, and identifies seven persistent research gaps that limit real-world deployment. Future enhancement pathways including IoT module integration, GPS-based geo-fencing, cloud fleet monitoring, EEG sensor fusion, and transformer-based attention architectures are discussed to guide the evolution of the field toward deployment-grade smart transportation systems.
Keywords: Driver Fatigue Detection; Eye Aspect Ratio (EAR); Mouth Aspect Ratio (MAR); Facial Landmark Detection; Computer Vision; OpenCV; Dlib; Drowsiness Monitoring; Real-Time Alert System; Convolutional Neural Network; LSTM; Smart Vehicles; Road Safety; Microsleep Detection; IoT Integration.
Keywords: Driver Fatigue Detection; Eye Aspect Ratio (EAR); Mouth Aspect Ratio (MAR); Facial Landmark Detection; Computer Vision; OpenCV; Dlib; Drowsiness Monitoring; Real-Time Alert System; Convolutional Neural Network; LSTM; Smart Vehicles; Road Safety; Microsleep Detection; IoT Integration.
Abstract
Towards Intelligent Healthcare: A Study of AI Applications, Challenges, and Future Trends
Kaveri Suresh Boraste, Akansha Gunvantrao Deshmukh, Kirti Dinkar More
DOI: 10.17148/IJARCCE.2026.155116
Abstract: Artificial Intelligence (AI) is rapidly reshaping modern healthcare systems by enabling intelligent data processing, improving diagnostic precision, and supporting clinical decision-making. With the growth of electronic health records (EHRs), medical imaging, and wearable technologies, healthcare institutions are generating vast amounts of heterogeneous data that require advanced analytical techniques. AI technologies, including machine learning, deep learning, and natural language processing, provide powerful tools to extract meaningful insights from such data. This paper presents a comprehensive and structured review of AI in healthcare, focusing on system architectures, core enabling technologies, real-world applications, and emerging challenges. It highlights recent advancements such as explainable AI, federated learning, and blockchain-integrated healthcare systems. Furthermore, critical concerns related to data privacy, bias, interpretability, and regulatory compliance are examined in detail. The study concludes that while AI significantly enhances healthcare efficiency, accuracy, and accessibility, its successful adoption depends on the development of transparent, ethical, and clinically validated systems.
Keywords: Artificial Intelligence, Healthcare Systems, Deep Learning, Clinical Decision Support, Explainable AI, Smart Healthcare
Keywords: Artificial Intelligence, Healthcare Systems, Deep Learning, Clinical Decision Support, Explainable AI, Smart Healthcare
Abstract
Research Design Approaches in MediNet – AI Health Risk and Smart Hospital Finder
Sainath Reddy Y S, Pani arvind, Sreenivasa Reddy, Sharath Kumar, Dr. Muhibur Rahman T.R
DOI: 10.17148/IJARCCE.2026.155117
Abstract: The majority of people do not seek medical attention until the condition becomes critical. This is because the usual health-related applications do not have anything innovative to offer beyond 5 basic symptom checkers. Moreover, they do not take into consideration the lifestyle of the user, like sleep patterns, diet, level of exercise, and stress levels, which are the actual causes of health risks. In the case of MediNet, the application of symptoms and daily routines is used to identify the potential risks of health complications, allowing the user to take necessary precautions rather than waiting for the condition to be critical. Moreover, this application also offers a feature that can be referred to as Smart Hospital Finder, which can automatically locate the nearest hospital that suits the health state of the user and provide the most convenient route to 2 that hospital via a map. In order to enable the user to track their health time, this application offers a feature that can be referred to as a Centralized Dashboard. The system is built using React.js, Node.js/Flask backend, MySQL database, and a Random Forestbased prediction model for health risk prediction, demonstrating that intelligent digital systems can meaningfully improve early diagnosis and reduce health risks.
Keywords: Artificial Intelligence, Health Risk Prediction, Random Forest, Smart Hospital Finder, Dijkstra's Algorithm, Preventive Healthcare, Machine Learning, Web-based Healthcare Platform, Lifestyle Analysis, Symptom Checker
Keywords: Artificial Intelligence, Health Risk Prediction, Random Forest, Smart Hospital Finder, Dijkstra's Algorithm, Preventive Healthcare, Machine Learning, Web-based Healthcare Platform, Lifestyle Analysis, Symptom Checker
Abstract
AI Powered Health and Automation Tools
Abhishek A, Prof. Vidya S
DOI: 10.17148/IJARCCE.2026.155118
Abstract: The rapid growth of digital platforms for communication, productivity, and health monitoring has led to fragmented workflows that require constant manual coordination. While existing automation tools reduce repetitive effort, they remain largely rule-based and lack contextual awareness, limiting their ability to adapt to unstructured information and human conditions. This paper presents AI-Powered Health and Automation Tools, an intelligent workflow orchestration platform that embeds large language models (LLMs) as native decision-making components and integrates real-time health data into automation logic. The system employs a visual, node-based workflow builder supported by a secure micro-kernel execution engine capable of cognitive branching and adaptive task execution. Health metrics such as sleep quality and activity levels dynamically influence workflow behaviour. Experimental evaluation demonstrates sub-100 ms local execution latency, high accuracy in health-triggered actions (94–97%), and strong usability among non-technical users. The results indicate that health-aware AI orchestration is both feasible and effective for next-generation automation systems.
Keywords: Artificial Intelligence, Workflow Orchestration, Large Language Models, Health Integration, Automation Platform
Keywords: Artificial Intelligence, Workflow Orchestration, Large Language Models, Health Integration, Automation Platform
Abstract
FitVision: Vision-Based Intelligent Fitness Assistance Using Pose-Guided Movement Analysis
Swati Uparkar, Purva Ambre, Aashna Anchan, Hasan Contractor, Husain Contractor
DOI: 10.17148/IJARCCE.2026.155119
Abstract: With the rise in home-based fitness activities, there is a growing need for systems that can guide fitness activities without the need for a trainer. In this paper, we propose a vision-based fitness assistance system called FitVision. The system includes real-time pose estimation, motion tracking over time, and a conversational interface. The system utilizes MediaPipe BlazePose for keypoint detection from video inputs. Based on keypoint detection, a rule-based approach involving joint angles is employed to interpret various exercises. Postures are also checked through joint angle calculations, while repetitions are tracked through a simple state-based approach. Besides the visual component, there is also a chatbot component known as FitBot, which assists users in their fitness-related queries. Moreover, the interaction becomes even more engaging. Based on our results, we observed that the system performs satisfactorily in identifying the exercises correctly, along with a reasonable repetition count. This shows that the use of computer vision along with a chatbot interface may be beneficial in creating a simple fitness assistant.
Keywords: Computer Vision, Pose Estimation, Fitness Monitoring, Deep Learning, Exercise Analysis
Keywords: Computer Vision, Pose Estimation, Fitness Monitoring, Deep Learning, Exercise Analysis
Abstract
Retrieval-Augmented Generation for Smarter Web Scraping and Synthesis: A Cache-Aware Multimodal Framework for Web Intelligence
Rishith Poojary, Aryan Rana, Aditya Magar, Dev Raval, Pravin Shinde
DOI: 10.17148/IJARCCE.2026.155120
Abstract: Most web pages today do not deliver data in straightforward ways. Content often appears only after scripts run, making early snapshots incomplete. Each visit might show a slightly altered layout. Identical details - like pricing or bylines - sit inside unpredictable tag arrangements depending on the site. Tools relying on fixed rules, such as CSS paths or XPath, function reliably until design changes occur. A minor update may disrupt what once worked without warning. Our system, WebRAG, takes another path altogether. Viewed as a form of information gathering, web scraping here builds on three distinct aspects per webpage: raw textual elements, structural markup shaped by creators, alongside how content visually appears during browsing - these together anchor a generation process firmly within up-to-date materials instead of outdated datasets. When assessing prior visits, a caching mechanism checks if underlying structures have shifted, skipping repeated processing where little or nothing differs from earlier versions. Each result includes documented origins detailing location online, specific document fragments involved, along with reliability estimates tied to recovery quality. Testing occurred using WebRAGBench - a collection exceeding five thousand hand-labeled examples drawn from news outlets, shopping platforms, and knowledge-focused websites. Performance surpassed rule- driven methods plus standard text-based pipelines when measuring correctness in fetching data, fidelity in pulling out answers, and overall response speed. A single 2.8× boost in throughput came just from caching. Where performance still lags becomes clear when examining system limits - this points toward promising paths forward.
Keywords: Retrieval-Augmented Generation, Web Scraping, Large Language Models, Multimodal Embeddings, Information Extraction
Keywords: Retrieval-Augmented Generation, Web Scraping, Large Language Models, Multimodal Embeddings, Information Extraction
Abstract
Engunity X : AI-Powered SaaS Platform for Research, Code Generation and Data Analysis
Shubh Rajesh Shah, Nigranth Shailesh Shah, Shreyash Chetan Mokani, Anurag Mrityunjay Kumar, Rahul Pachade
DOI: 10.17148/IJARCCE.2026.155121
Abstract: Engineering students routinely juggle five or more disconnected tools in a single work session—a paper search engine, a chatbot, a code editor, a sandbox runtime, and some method for pressure-testing assumptions—and every context switch between them discards the mental state built up in the one just left. This paper presents Engunity X, a unified five-service SaaS platform that collapses that workflow into one shared-memory environment. The centrepiece is OmniRAG, a complexity-adaptive retrieval pipeline backed by a fine-tuned DistilBERT classifier (18,500 labelled queries, 91.3 % macro-F1) that routes each request to the most appropriate of four strategies: direct generation, hybrid dense-sparse retrieval, knowledge-graph traversal, or recursive chain-of-thought. On a 500-query benchmark drawn from publicly available CS documentation, OmniRAG reached 88.0 % retrieval accuracy on multi- hop questions—24 percentage points above a standard single-strategy FAISS baseline. A Docker-sandboxed Code Lab corrected 70.7 % of seeded program errors autonomously in one iteration. An adversarial Decision Vault flagged 78 % of logically weak arguments against a human-rated gold set. The full stack sustained P95 latency below 500 ms at 500 concurrent users on commodity hardware.
Keywords: retrieval-augmented generation; autonomous agents; knowledge graph; secure sandbox; adversarial reasoning; DistilBERT; engineering education; SaaS.
Keywords: retrieval-augmented generation; autonomous agents; knowledge graph; secure sandbox; adversarial reasoning; DistilBERT; engineering education; SaaS.
Abstract
DermAI: A Vision Transformer-Based Web System for Multi-Class Skin Disease Detection
Utilizing DINOv, Priyanka Dnyaneshwar Muley
DOI: 10.17148/IJARCCE.2026.155122
Abstract: Dermatological conditions represent a significant global health burden, often requiring specialized expertise for accurate diagnosis. Early detection of skin diseases, particularly malignancies like melanoma, is critical for improving patient outcomes. This paper presents DermAI, an advanced, production-ready skin disease detection system powered by a fine-tuned DINOv2-Base Vision Transformer (ViT) architecture. By leveraging self-supervised features distilled from large-scale image data, DermAI achieves a validation accuracy of 95.57% across 31 distinct skin disease classes. The proposed system integrates a high-performance deep learning backend with a user-friendly web interface, facilitating real-time inference and ranking predictions by confidence. Our results demonstrate that foundational vision models can be effectively repurposed for specialized medical diagnostic tasks with high precision and reliability. The global prevalence of dermatological conditions, coupled with a shortage of specialized dermatologists, necessitates the development of accessible, highly accurate automated diagnostic tools. This paper presents DermAI, an end-to-end web- based system designed for the classification of 31 distinct skin diseases. By integrating a fine-tuned DINOv2-Base Vision Transformer (ViT-B/14) with a robust full-stack web architecture (Python, Flask, HTML, CSS), the system delivers real- time, confidence-ranked predictions to the end-user. The underlying deep learning model -images, achieving a remarkable validation accuracy of 95.57% over 10 epochs. Through detailed case studies—including the high-confidence identification of malignant melanoma and benign fungal infections—this paper demonstrates that foundational self- supervised models can be effectively deployed via lightweight web frameworks to serve as reliable clinical decision- support systems.
Abstract
AI-Assisted Computer-Aided Diagnosis for Early Detection of Lung Cancer
Ghanashyam Pawar, Pranav Shinde, Bharati Mahale
DOI: 10.17148/IJARCCE.2026.155123
Abstract: Lung cancer is one of the major causes of death across the world, and early detection plays an important role in improving patient survival and treatment success. Computer-aided diagnosis (CAD) systems using CT scan images help doctors identify and classify lung nodules more effectively, supporting early-stage lung cancer diagnosis. Earlier machine learning methods such as Support Vector Machine (SVM) and K-Nearest Neighbor (KNN) were widely used, but they faced difficulties in analyzing large and complex medical image data.
With the growth of deep learning, medical image analysis has improved significantly. Advanced techniques such as Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and Generative Adversarial Networks (GAN) provide better accuracy in detecting, segmenting, and classifying pulmonary nodules. This review highlights recent developments in deep learning approaches for lung cancer diagnosis and explains how these methods perform better than traditional machine learning techniques.
The study also discusses the use of ensemble models and other modern approaches that increase the reliability and efficiency of pulmonary nodule analysis. Overall, deep learning has shown great potential in improving the accuracy of lung cancer detection and diagnosis. Despite some existing challenges, ongoing advancements in artificial intelligence are expected to further enhance early diagnosis and medical decision-making in the future.
Keywords: Lung cancer, Artificial Intelligence, Computer -aided diagnosis(CAD), Pumlonary nodule segmentation and classification, Deep Learning
With the growth of deep learning, medical image analysis has improved significantly. Advanced techniques such as Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and Generative Adversarial Networks (GAN) provide better accuracy in detecting, segmenting, and classifying pulmonary nodules. This review highlights recent developments in deep learning approaches for lung cancer diagnosis and explains how these methods perform better than traditional machine learning techniques.
The study also discusses the use of ensemble models and other modern approaches that increase the reliability and efficiency of pulmonary nodule analysis. Overall, deep learning has shown great potential in improving the accuracy of lung cancer detection and diagnosis. Despite some existing challenges, ongoing advancements in artificial intelligence are expected to further enhance early diagnosis and medical decision-making in the future.
Keywords: Lung cancer, Artificial Intelligence, Computer -aided diagnosis(CAD), Pumlonary nodule segmentation and classification, Deep Learning
Abstract
AI and its Impact on Human Emotion
Rajesh K N, Vidya S
DOI: 10.17148/IJARCCE.2026.155124
Abstract: Artificial Intelligence (AI) has increasingly become part of everyday human interaction, influencing not only technical operations but also emotions, behavior, decision-making, and interpersonal relationships. Advancements in conversational AI, recommendation systems, and emotionally responsive technologies have encouraged users to rely on AI-generated responses for emotional support, personal guidance, and decision assistance. This study examines the impact of AI on human emotions, particularly in the areas of decision-making, emotional dependency, and relationship dynamics. The research adopted a qualitative review-based methodology involving the analysis of 18 research papers published between 2018 and 2026 related to Artificial Intelligence, Human–AI Interaction, affective computing, and emotional AI systems. The study critically analyzed major AI models and algorithms including Transformer architectures, Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) networks, Reinforcement Learning systems, and recommendation algorithms to understand their role in emotionally adaptive interactions. The findings revealed that AI systems increasingly influence emotional reasoning and behavioral patterns through personalized responses, emotional simulation, and algorithmic recommendations. However, the study also identified significant limitations, including dependence on incomplete user inputs, lack of genuine emotional intelligence, probabilistic reasoning, and inability to fully understand complex human emotions and contextual realities. Furthermore, excessive dependence on AI systems may contribute to emotional bias reinforcement, cognitive dependency, and reduced independent critical thinking. The study concludes that while AI technologies offer convenience, accessibility, and emotional support, future development must prioritize human-centered AI design, ethical governance, emotional transparency, and preservation of human emotional autonomy to ensure balanced and responsible Human–AI Interaction.
Keywords: Artificial Intelligence, Human Emotions, Human–AI Interaction, Conversational AI, Emotional Dependency, Decision-Making, Recommendation Systems, Emotional AI, Affective Computing, Ethical AI
Keywords: Artificial Intelligence, Human Emotions, Human–AI Interaction, Conversational AI, Emotional Dependency, Decision-Making, Recommendation Systems, Emotional AI, Affective Computing, Ethical AI
Abstract
Copyright Issues in Digital Content Platforms
MOHAMED ZAID, A SAI SANDEEP, Dr. Muhibur Rahaman T.R
DOI: 10.17148/IJARCCE.2026.155125
Abstract: The exponential growth of digital content platforms such as streaming services, social media networks, and user-generated content portals has fundamentally transformed how creative works are distributed, consumed, and monetized. While these platforms have democratized access to information and creative content, they have simultaneously introduced complex copyright challenges that existing legal frameworks struggle to address. Issues such as unauthorized reproduction, algorithmic content detection failures, cross-border jurisdictional conflicts, and the ambiguous ownership of AI-generated content are increasingly prevalent. This paper examines the key copyright concerns arising within digital content ecosystems, analyses existing enforcement mechanisms including the Digital Millennium Copyright Act (DMCA) and the European Union Copyright Directive (EUCD), and evaluates the effectiveness of platform-level tools such as Content ID systems. A structured four-tier taxonomy of copyright enforcement approaches is proposed, ranging from manual reporting to intelligent AI-driven detection. The paper also identifies critical research gaps and concludes with recommendations for a balanced, scalable, and legally coherent framework for copyright governance in the digital era.
Abstract
INTELLIGENT PRIVACY PRESERVING DATA ENCRYPTION AND AΝΟΝΥΜΙΖΑΤΙON SYSTEM
Sangeetha M.E., Sanjeevi R B, Vinoth M
DOI: 10.17148/IJARCCE.2026.155126
Abstract: The project not only addresses privacy and security needs but also the issue of trust in data sharing practices. By providing a privacy-preserving system that guarantees the confidentiality of individuals' information, the project aims at creating a secure and reliable environment for data sharing. Additionally, the proposed system will support the implementation of various use cases such as health care, banking, and e-commerce where data privacy and security are of paramount importance. The integration of various advanced techniques like decision tree, random forest, or support vector machine for data sensitivity assessment and the selection of the most suitable cryptographic technique for the different types of data to be processed will be also feasible. This will help in providing the best possible solution in terms of data privacy and accessibility.Ultimately, the project is expected to make a strong contribution to the development of privacy-preserving technologies through its innovative approach and the development of the proposed Intelligent Privacy Preserving Data Encryption and Anonymization System. In the long run, the project's results might have a significant impact on the future of data protection technologies as more and more organizations migrate to cloud services and share information across borders.
Keywords: Data protection, encryption, anonymization, privacy-preserving, cloud computing, big data analytics, interconnected systems, decision making, automated systems, trust
Keywords: Data protection, encryption, anonymization, privacy-preserving, cloud computing, big data analytics, interconnected systems, decision making, automated systems, trust
Abstract
Krishi Mitra: A Multilingual AI-Powered Conversational Agent for Indian Farmers Integrating Government Schemes, Real-Time Mandi Prices, Crop Disease Detection, and Agricultural Advisories.
Jayashri Gajre, Sujay Manasubdar, Ishan Chaudhari, Shubham Panigrahi, Suraj Patil
DOI: 10.17148/IJARCCE.2026.155127
Abstract: Agriculture remains the backbone of India's economy, employing approximately 55% of the total workforce. Yet the majority of Indian farmers – particularly those in rural areas – continue to face significant challenges in accessing timely, accurate, and language-appropriate information about government welfare schemes, crop insurance policies, real- time market prices, and crop disease management. Existing digital solutions are fragmented, language-exclusive (primarily English or Hindi), and fail to deliver the holistic advisory support that smallholder farmers require. This paper presents Krishi Mitra ("Farmer's Friend"), a multilingual, AI-powered conversational chatbot designed specifically to bridge this information gap. The system integrates a Retrieval-Augmented Generation (RAG) pipeline using ChromaDB as a vector store and the paraphrase-multilingual-MiniLM-L12-v2 sentence-transformer model for semantic embedding. Natural language responses are generated using the Llama-3.3-70b-versatile model served through the Groq API. The chatbot supports three languages – English, Hindi, and Marathi – with automatic language detection and cross-lingual translation performed via LLM prompting. Additionally, the system incorporates a real-time mandi (agricultural market) price retrieval module fetching live data from government portals, a Convolutional Neural Network (CNN)-based crop disease prediction sub-system using leaf image classification, and a Text-to-Speech (TTS) module powered by Google Text-to-Speech (gTTS) for audio accessibility. Deployed as a Flask REST API with a React.js frontend, the system was evaluated across multiple agricultural query categories, demonstrating high semantic relevance, language fidelity, and user accessibility.
Keywords: Artificial Intelligence, Chatbot, Retrieval-Augmented Generation, Multilingual NLP, Indian Agriculture, Farmer Advisory, ChromaDB, Crop Disease Detection, Mandi Prices, Sentence Transformers, Groq, LLM.
Keywords: Artificial Intelligence, Chatbot, Retrieval-Augmented Generation, Multilingual NLP, Indian Agriculture, Farmer Advisory, ChromaDB, Crop Disease Detection, Mandi Prices, Sentence Transformers, Groq, LLM.
Abstract
AI POWERED REAL ESTATE SYSTEM
Abhishek, Deepika V M
DOI: 10.17148/IJARCCE.2026.155128
Abstract: The management and discovery of real estate properties is a critical task in modern property platforms, requiring efficient and accurate methods to assist users in making informed decisions. Traditional real estate systems often rely on manual browsing and basic filtering, which can be time-consuming and less personalized. In recent years, artificial intelligence–based approaches have emerged as powerful tools for enhancing property search and management systems. This paper presents a study on the application of an AI-powered real estate system for intelligent property listing, search, and appointment management. The proposed system leverages modern web technologies and AI-assisted filtering mechanisms to improve the accuracy and relevance of property recommendations. The system architecture follows a modular design where the backend handles data processing and storage, while the frontend provides an interactive and user-friendly interface. Intelligent filtering and sorting mechanisms help users efficiently locate properties based on preferences such as location, price, and property type. The results demonstrate that the AI-powered real estate system improves user experience by providing faster property discovery and efficient management compared to traditional listing platforms. The application is implemented as a web-based system using React.js on the frontend with a Node.js and Express backend, and MongoDB for data storage.
Keywords: Artificial Intelligence (AI), Real Estate Management, Property Listing, Node.js, Express.js, React.js, MongoDB, JWT Authentication, Appointment Scheduling, Admin Dashboard, Web Application, Property Recommendation, Role-Based Access Control, ImageKit, Full-Stack Development, Data Analytics, User Interaction Tracking, Property Search, Intelligent Filtering, Decision Support System.
Keywords: Artificial Intelligence (AI), Real Estate Management, Property Listing, Node.js, Express.js, React.js, MongoDB, JWT Authentication, Appointment Scheduling, Admin Dashboard, Web Application, Property Recommendation, Role-Based Access Control, ImageKit, Full-Stack Development, Data Analytics, User Interaction Tracking, Property Search, Intelligent Filtering, Decision Support System.
Abstract
HEPATOSCAN:AI BASED LIVER TUMOR SEGMENTATION SYSTEM
Deekshith B N, Theerthashree G S
DOI: 10.17148/IJARCCE.2026.155129
Abstract: Liver cancer is one of the leading causes of cancer-related deaths worldwide, and early diagnosis plays a crucial role in improving treatment outcomes and patient survival. Traditional methods of analyzing CT scan images rely heavily on manual interpretation by radiologists, which can be time-consuming, labor-intensive, and prone to human error. To address these challenges, this project presents HepatoScan, an AI-based web application designed for automated liver and tumor segmentation using deep learning techniques.
The proposed system utilizes a U-Net–based convolutional neural network implemented using TensorFlow to perform accurate pixel-level segmentation of liver and tumor regions from CT scan images. Prior to segmentation, the images undergo preprocessing steps such as grayscale conversion, resizing, and normalization to ensure consistent input quality for the model. The trained model generates liver and tumor masks, enabling automated identification of affected regions. In addition to segmentation, the system performs tumor size estimation using pixel-based area calculation and provides a user-friendly web interface developed using Flask and web technologies.
The application also includes automated PDF medical report generation containing the uploaded CT image, segmentation outputs, and tumor size details. By integrating deep learning, medical image processing, and automated reporting into a single platform, HepatoScan aims to improve diagnostic efficiency, reduce manual effort, and support radiologists in clinical decision-making.
Keywords: Artificial Intelligence (AI), Deep Learning, Medical Image Segmentation, Liver Tumor Detection, U-Net Architecture, Convolutional Neural Network (CNN), TensorFlow, CT Scan Analysis, Flask Web Application, Tumor Size Estimation, Image Preprocessing, Medical Imaging, Computer Vision, Automated PDF Report Generation, Healthcare AI, Liver Cancer Diagnosis, Supervised Learning, Biomedical Image Analysis.
The proposed system utilizes a U-Net–based convolutional neural network implemented using TensorFlow to perform accurate pixel-level segmentation of liver and tumor regions from CT scan images. Prior to segmentation, the images undergo preprocessing steps such as grayscale conversion, resizing, and normalization to ensure consistent input quality for the model. The trained model generates liver and tumor masks, enabling automated identification of affected regions. In addition to segmentation, the system performs tumor size estimation using pixel-based area calculation and provides a user-friendly web interface developed using Flask and web technologies.
The application also includes automated PDF medical report generation containing the uploaded CT image, segmentation outputs, and tumor size details. By integrating deep learning, medical image processing, and automated reporting into a single platform, HepatoScan aims to improve diagnostic efficiency, reduce manual effort, and support radiologists in clinical decision-making.
Keywords: Artificial Intelligence (AI), Deep Learning, Medical Image Segmentation, Liver Tumor Detection, U-Net Architecture, Convolutional Neural Network (CNN), TensorFlow, CT Scan Analysis, Flask Web Application, Tumor Size Estimation, Image Preprocessing, Medical Imaging, Computer Vision, Automated PDF Report Generation, Healthcare AI, Liver Cancer Diagnosis, Supervised Learning, Biomedical Image Analysis.
Abstract
From Paper to Prototype: An End-to-End AI System for Automated Research Implementation
Taranjeet Singh, Monish Patil, Tejas Mungekar, Swaraj Gadre, Rupali Shinde
DOI: 10.17148/IJARCCE.2026.155130
Abstract: Bridging the gap between academic research and real-world implementation remains a complex challenge in the technological ecosystem. While research papers propose algorithms and frameworks, converting them into runnable prototypes requires significant time, effort, and programming knowledge. This paper presents Research to Reality (PROTOGEN), an intelligent, end-to-end AI-driven system that automates the translation of academic research into executable software. By leveraging Natural Language Processing (NLP), a multi-model Large Language Model (LLM) pipeline powered by NVIDIA NIM, Groq (LLaMA-3.3-70B), and Google Gemini 2.0 Flash, the system performs text extraction, summarization, ideation, code generation, and live execution via Docker containerization. The system achieves a prototype generation success rate of 78.4% across 50 research papers spanning 6 domains, with an average end-to-end pipeline latency of 42 seconds. The work combines theoretical understanding with practical implementation, illustrating how AI systems can accelerate innovation, enhance collaboration, and make research outputs more actionable.
Keywords: AI code generation, NLP, LLM, software prototyping, research automation, human-in-the-loop, program synthesis, Docker containerization, NVIDIA NIM, multi-model pipeline .
Keywords: AI code generation, NLP, LLM, software prototyping, research automation, human-in-the-loop, program synthesis, Docker containerization, NVIDIA NIM, multi-model pipeline .
Abstract
MedAI-DX: An AI-Powered Real-Time Clinical Decision Support System for Resource-Constrained Healthcare Environments
Hemanth Gowda A, Jayanth Somashekar, Akarsh M, Vinay Gowda PN, Dr. Kavitha AS
DOI: 10.17148/IJARCCE.2026.155131
Abstract: Rural and peri-urban healthcare facilities in developing nations face a critical physician shortage, with a single clinician often managing 80–120 patients daily under severe diagnostic resource constraints. Existing clinical decision support systems are predominantly cloud-only, English-exclusive, unimodal, or cost-prohibitive for deployment at primary health centres. This paper presents MedAI-DX, a multimodal, real-time AI-powered clinical decision support platform designed specifically for resource-constrained environments. The proposed system integrates a fine-tuned clinical Natural Language Processing (NLP) engine based on BioMistral-7B for multilingual symptom extraction and ICD-11 mapping, an EfficientNet-B4 computer vision module trained on NIH ChestX-ray14, ISIC 2020, and APTOS 2019 datasets for diagnostic image classification with Grad-CAM explainability, and a weighted evidence fusion risk stratification engine producing Green/Amber/Red triage classifications with structured referral recommendations. The system achieves a clinical entity F1 score of 0.84, image classification AUC-ROC of 0.87 across four disease categories, and an end-to-end latency under 2.5 seconds on standard cloud infrastructure. The full-stack deployment — React frontend, FastAPI backend, PostgreSQL with pgvector — is validated through a live interactive demonstration accessible via web browser. This work contributes a comprehensive, ethically grounded framework for AI-augmented clinical decision-making in multilingual, low-resource settings.
Keywords: Clinical decision support, multimodal AI, natural language processing, computer vision, risk stratification, ICD-11, Grad-CAM, EfficientNet, BioMistral, FastAPI, pgvector, healthcare AI, triage, multilingual NLP.
Keywords: Clinical decision support, multimodal AI, natural language processing, computer vision, risk stratification, ICD-11, Grad-CAM, EfficientNet, BioMistral, FastAPI, pgvector, healthcare AI, triage, multilingual NLP.
Abstract
Early Detection of Comorbid Anxiety and Depression Using Explainable Machine Learning on DASS-21 Psychometric Data
Pranto Bosu, Satinder Kaur, Tajbir Singh
DOI: 10.17148/IJARCCE.2026.155132
Abstract: Depression and anxiety frequently co-occur, yet early detection of their comorbidity remains challenging due to reliance on subjective clinical assessments. This study presents an explainable machine learning framework for binary classification of joint anxiety-depression at-risk status using the DASS-21 psychometric questionnaire. To prevent data leakage, we employ a stress-proxy feature strategy that excludes depression and anxiety subscale items from the input features, retaining only stress-related questionnaire items and demographic variables. Six classifiers—Logistic Regression, SVM, Random Forest, XGBoost, Gradient Boosting, and an MLP neural network—are evaluated using 5- fold stratified cross-validation with SMOTE-based class balancing. The best-performing model, Random Forest (tuned), achieves a test accuracy of 60.10% and ROC-AUC of 0.4917 under the leakage-free setting, highlighting the inherent difficulty of predicting comorbid risk from indirect indicators alone. SHAP (SHapley Additive exPlanations) analysis identifies education level and DASS-21 item Q1A (difficulty winding down) as the most influential predictors. Demographic fairness analysis reveals comparable performance across gender and age subgroups. These findings establish a transparent, reproducible baseline for comorbid mental health screening and underscore the need for richer multi-modal feature sets to improve predictive accuracy.
Keywords: machine learning; explainable AI; depression; anxiety; comorbidity; DASS-21; SHAP; mental health screening
Keywords: machine learning; explainable AI; depression; anxiety; comorbidity; DASS-21; SHAP; mental health screening
Abstract
Distributed Denial of Service Attack Detection using Machine Learning
Namrata Sunil bodhale
DOI: 10.17148/IJARCCE.2026.155133
Abstract: The rapid growth of 5G communication and Internet of Things (IoT) devices has significantly increased the risk of Distributed Denial of Service (DDoS) attacks in modern networks. These attacks attempt to interrupt network services by flooding systems with excessive traffic, thereby affecting availability, reliability, and performance. Traditional detection mechanisms are often unable to identify evolving attack patterns efficiently, especially in high- speed 5G environments.
This research presents a machine learning-based framework for detecting DDoS attacks in 5G-enabled IoT networks. The proposed system utilizes Artificial Neural Networks (ANN) with Bayesian Regularization and backpropagation techniques to classify malicious and normal traffic. The model performs preprocessing, feature selection, training, and validation using network traffic datasets. The proposed framework improves detection accuracy while reducing false- positive rates. Experimental analysis demonstrates that machine learning methods can effectively identify abnormal traffic behavior and support real-time network protection mechanisms.
Keywords: DDoS Attack, Machine Learning, Artificial Neural Network, 5G Networks, IoT Security, Bayesian Regularization.
This research presents a machine learning-based framework for detecting DDoS attacks in 5G-enabled IoT networks. The proposed system utilizes Artificial Neural Networks (ANN) with Bayesian Regularization and backpropagation techniques to classify malicious and normal traffic. The model performs preprocessing, feature selection, training, and validation using network traffic datasets. The proposed framework improves detection accuracy while reducing false- positive rates. Experimental analysis demonstrates that machine learning methods can effectively identify abnormal traffic behavior and support real-time network protection mechanisms.
Keywords: DDoS Attack, Machine Learning, Artificial Neural Network, 5G Networks, IoT Security, Bayesian Regularization.
Abstract
VERIPILL:ML BASED COUNTERFEIT MEDICINE PREDICTION
Mrs. Archana N, Tarun R, Monika. K, Pranav Ramesh, Priya R K
DOI: 10.17148/IJARCCE.2026.155134
Abstract: Counterfeit medicines pose a major health risk in World, where visually identical fake drugs often reach consumers without detection. VeriPill is an ML-based platform designed to help users verify medicine authenticity through image analysis. The system combines Convolutional Neural Networks (CNN), Optical Character Recognition (OCR), and dataset validation to examine packaging features and extracted label details, identifying tampering and inconsistencies that indicate counterfeit products. Alongside verification, VeriPill offers a symptom-based medicine guide and a nearby pharmacy locator for added usability. Built using Python, Django, and OpenCV, the platform provides fast and accessible drug authentication, supporting safer medicine use and strengthening trust in the pharmaceutical ecosystem.
Keywords: counterfeit detection, CNN, OCR, computer vision, medicine authentication, AI in healthcare.
Keywords: counterfeit detection, CNN, OCR, computer vision, medicine authentication, AI in healthcare.
Abstract
Adaptive SPD-YOLO: Enhancing Spatial Feature Retention for Lunar Boulder Detection
Dhruv Solanki, Shashank Singh, Shrawani Sawant, Sudhanshu Singh, Ms. Swati Uparkar
DOI: 10.17148/IJARCCE.2026.155135
Abstract: Mainstream real-time object detectors routinely sacrifice spatial resolution to keep inference costs manageable, a trade-off that proves especially damaging when the targets of interest span only a handful of pixels. This work introduces AdaptiveSPD-YOLO, a modified YOLOv26 architecture that counters this loss through an Adaptive Space-to-Depth downsampling module inserted at the P3/8 backbone junction. Rather than indiscriminately rearranging every channel into the depth dimension, the module employs a Squeeze-and-Excitation–style channel-attention gate that scores each feature map by its informational salience before the spatial-to-channel rearrangement takes place. To quantify the benefits of this selective preservation strategy, a variance-based Spatial Retention Tracking protocol is introduced and monitored across training epochs. Experiments on a large-scale lunar boulder dataset comprising 23,154 multiple scale orbital images with 8,94,474 annotated bounding boxes yield a peak mAP@50 of 78.1% and a precision of 76.8% at a computational cost of 70.6 GFLOPs. Ablation analysis confirms that the attention-gated variant surpasses both standard strided convolution and uniform SPD, while the channel-attention gate autonomously increases its suppression rate from 49.9% to 61.8% during training, indicating an emergent capacity for discriminative feature selection that directly correlates with improved detection accuracy.
Keywords: Adaptive Space-to-Depth, YOLOv26, Channel Attention, Spatial Information Retention, Lunar Boulder Detection, Small Object Detection
Keywords: Adaptive Space-to-Depth, YOLOv26, Channel Attention, Spatial Information Retention, Lunar Boulder Detection, Small Object Detection
Abstract
A SecureStep-Smart Personal Security and Safe Navigation Platform
Dhanalakshmi S, Prajwal B, Poonamlal, Rajeev BS, Pavan Sai
DOI: 10.17148/IJARCCE.2026.155136
Abstract: Personal safety is becoming a bigger concern these days. you. Most apps only use SOS messages and location sharing, and ignore the bigger picture. They can’t identify As crime increases and emergency services take longer to respond, people feel more vulnerable, especially when they are traveling. Safety apps often only let users send an alert or share their location, and they stop there, without further features. New technologies like Artificial Intelligence (AI), Internet of Things (IoT), GPS, cloud computing, and smartphones have enabled the creation of security platforms that offer more than just basic alerts. These systems can now monitor your activity, look around you, and help during emergencies in real time. This survey looks at what researchers have done so far with smart personal safety systems and safe navigation tools. It covers everything from AI- powered monitoring and IoT devices to GPS trackers, wearables, and innovative uses of crime data to plan safer routes. It also examines the pros and cons of each method and the major challenges that still remain. Putting AI at the center, along with advanced navigation, significantly improves personal safety. SecureStep brings all these features together under one platform: real- time monitoring, emergency communication, and safe route guidance. It represents a step toward greater reliability and a stronger sense of security.
Keywords: Personal Safety, Smart Navigation, Artificial Intelligence, IoT, GPS Tracking, Emergency Response, Threat Detection, Mobile Security System, Secure Routing, Real-Time Monitoring.
Keywords: Personal Safety, Smart Navigation, Artificial Intelligence, IoT, GPS Tracking, Emergency Response, Threat Detection, Mobile Security System, Secure Routing, Real-Time Monitoring.
Abstract
AI-Based Early Detection and Risk Prediction of Jaundice Using Clinical and Liver Function Test Data
Vaseekaran A, Hariharan S, and Mrs. Malathi G
DOI: 10.17148/IJARCCE.2026.155137
Abstract: Jaundice, the yellow discolouration of skin and sclera caused by elevated serum bilirubin, is a clinically visible manifestation of hepatic, haematological or biliary dysfunction whose early detection materially affects patient outcomes. In the Indian subcontinent, where viral hepatitis A and E are endemic and non-alcoholic fatty liver disease is rising in prevalence, the burden of liver disease is substantial while specialist hepatology expertise is concentrated in tier-one urban centres. This paper presents HepatIQ, an artificial-intelligence-driven decision support system for the early detection and risk stratification of jaundice. The system combines a Random Forest classifier trained on the Indian Liver Patient Dataset of 583 records with a rule-based pattern classifier and a biochemical flagging engine to produce explainable risk assessments. The system achieves a five-fold cross-validation accuracy of 70.68% (±3.14%) and a test- set accuracy of 72.65%, with a clinically critical recall of 100% on the high-risk class at 97% precision. The hybrid combination of probabilistic machine learning output, pattern classification, biochemical flags and India-specific dietary recommendations addresses the explainability gap that limits adoption of black-box predictors in clinical settings. The complete system is delivered as a Flask-based web application with SQLite persistence, runs on commodity hardware, and uses only open-source libraries.
Keywords: Jaundice, Liver Function Test, Random Forest, Machine Learning, Indian Liver Patient Dataset, Risk Stratification, Clinical Decision Support, Explainable AI, Hepatology, Bilirubin.
Keywords: Jaundice, Liver Function Test, Random Forest, Machine Learning, Indian Liver Patient Dataset, Risk Stratification, Clinical Decision Support, Explainable AI, Hepatology, Bilirubin.
Abstract
Smartbill Intelligent Retail Invoice And Profit Analysis System
P.R Ajitha M.E, MBA, Gokul P, Arvinth Kumar A
DOI: 10.17148/IJARCCE.2026.155138
Abstract: Retail businesses need efficient billing and accurate profit tracking to stay competitive in today's changing market. Many small and medium retailers still use traditional billing systems that only generate invoices and do not provide smart profit analysis. SMARTBILL, an Intelligent Retail Invoice and Profit Analysis System, automates invoice processing and offers real-time profit calculation and sales analytics. The system brings together inventory management, cost tracking, tax calculation, and dynamic profit margin analysis in one platform. By using database-driven design and analytical algorithms, SMARTBILL helps retailers track revenue trends, find high-performing products, and make informed business choices. This system improves operational efficiency, cuts down on manual errors, increases financial transparency, and aids in planning for sustainable retail growth.
Keywords: Invoice Automation, Profit Analysis, Inventory Management, Sales Analytics, Revenue Tracking, Business Intelligence, Financial Transparency, Retail Data Analytics.
Keywords: Invoice Automation, Profit Analysis, Inventory Management, Sales Analytics, Revenue Tracking, Business Intelligence, Financial Transparency, Retail Data Analytics.
Abstract
A Lightweight Wireless Intrusion Detection System for Real-Time Deauthentication and Rogue Access Point Mitigation.
Nandini Rajesh Kasar
DOI: 10.17148/IJARCCE.2026.155139
Abstract: The pervasive deployment of IEEE 802.11 wireless networks has revolutionized digital connectivity, but it has simultaneously expanded the attack surface for threat actors. A critical vulnerability within legacy Wi-Fi protocols is the transmission of management frames in an unencrypted and unauthenticated format. This flaw is routinely exploited to execute deauthentication denial-of-service (DoS) attacks and deploy Rogue Access Points (RAPs) or "Evil Twins" to intercept sensitive data. This paper presents a lightweight, low-cost Wireless Intrusion Detection System (WIDS) architecture utilizing a Raspberry Pi and monitor-mode network adapters. By employing a Python-based detection engine leveraging Scapy and tcpdump, the proposed system effectively identifies deauthentication floods, SSID-BSSID duplication, MAC spoofing, and abnormal Received Signal Strength Indicator (RSSI) variations. Experimental results validate the system's ability to provide real-time alerts and forensic packet captures (PCAP) with minimal computational overhead, offering a highly scalable solution for enterprise and edge-network security.
Abstract
AI-POWERED SMART INTERVIEW ASSISTANT WITH REAL-TIME FEEDBACK AND SCORE PREDICTION
Dr. Umma Habiba, Jeni Pragashini J, Kayathri A, Yogalakshmi M
DOI: 10.17148/IJARCCE.2026.155140
Abstract: The rapid growth of Artificial Intelligence has created new opportunities to improve traditional interview preparation methods. Many students struggle with confidence, communication and evaluating the quality of their responses during job interviews. This paper proposes an AI-Powered Smart Interview Assistant that evaluates candidate responses and provides real-time feedback and score prediction using machine learning and natural language processing techniques. The proposed system allows candidates to answer interview questions through text or voice input. The system analyzes the responses using natural language processing methods to evaluate relevance, grammar, semantic similarity and communication quality. A machine learning model then predicts the interview score based on extracted linguistic and contextual features. The system was tested with different categories of answers including relevant, irrelevant, short and filler-based responses. Experimental results show that the proposed system achieves model accuracy between 88 percent and 92 percent while maintaining an average response time of less than three - five seconds. The speech recognition component achieves approximately 90 percent accuracy in converting spoken responses into text. The system provides immediate suggestions that help candidates improve communication skills, reduce interview anxiety and enhance overall interview readiness. The proposed framework demonstrates the potential of artificial intelligence in building intelligent training tools that support career preparation and professional development.
Keywords: Artificial Intelligence, Interview Assistant, Natural Language Processing, Machine Learning, Speech Recognition, Score Prediction
Abstract
AI-Powered Automated Business Report Generator
Dr. T. R. Muhibur Rahman, Haseeba Kouser, Nazia Taj S, Sai Vaishnavi D
DOI: 10.17148/IJARCCE.2026.155141
Abstract: The integration of Artificial Intelligence (AI) into automated business report generation has significantly transformed data-driven decision-making by minimizing manual effort and improving analytical precision. The AI- Powered Automated Business Report Generator provides an intelligent framework for extracting, analysing, and visualizing data from heterogeneous sources, including PDF, Excel, and text files. It incorporates Natural Language Processing (NLP) and machine learning techniques to automatically extract insights, generate summaries, and perform sentiment analysis using models such as T5 and DistilBERT. The system dynamically identifies data types, processes structured and unstructured information, and produces interactive visualizations including bar charts, scatter plots, histograms, and word clouds. The output is compiled into a comprehensive digital report that combines textual interpretation and visual analytics, offering a streamlined solution for real-time business intelligence. This automation ensures accuracy, scalability, and faster insight generation, significantly reducing human workload in report preparation.
Keywords: Artificial Intelligence, Business Report Generation, Natural Language Processing, T5, DistilBERT, Data Visualization.
Keywords: Artificial Intelligence, Business Report Generation, Natural Language Processing, T5, DistilBERT, Data Visualization.
Abstract
HIDDEN IN PLAIN SIGHT: MODERN TECHNIQUES FOR PII PROTECTION
Shashank M N, Sandarsh Gowda MM
DOI: 10.17148/IJARCCE.2026.155142
Abstract: In the modern digital economy, organizations increasingly rely on Business-to-Business (B2B) platforms to process and exchange large volumes of sensitive user information. Personally Identifiable Information (PII) such as phone numbers, email addresses, identity numbers, and financial records forms the backbone of many digital services. However, the increasing frequency of cyberattacks and data breaches has exposed the limitations of traditional security approaches that rely primarily on network perimeters and access control mechanisms. Once attackers bypass these defenses, sensitive information stored in databases becomes vulnerable.
Modern data protection strategies therefore emphasize encryption-based security, where the data itself remains protected even if unauthorized access occurs. This research investigates advanced encryption mechanisms designed specifically for B2B systems that must balance security, usability, and performance. The study focuses on techniques such as Format- Preserving Encryption (FPE), Searchable Symmetric Encryption (SSE), and Privacy Enhancing Technologies (PETs), which allow organizations to store and process encrypted data without disrupting existing system architectures.
Keywords: PII Encryption, Format-Preserving Encryption (FPE), Searchable Symmetric Encryption (SSE), B2B Data Security, DPDP Act 2025, Privacy-Enhancing Technologies (PETs).
Modern data protection strategies therefore emphasize encryption-based security, where the data itself remains protected even if unauthorized access occurs. This research investigates advanced encryption mechanisms designed specifically for B2B systems that must balance security, usability, and performance. The study focuses on techniques such as Format- Preserving Encryption (FPE), Searchable Symmetric Encryption (SSE), and Privacy Enhancing Technologies (PETs), which allow organizations to store and process encrypted data without disrupting existing system architectures.
Keywords: PII Encryption, Format-Preserving Encryption (FPE), Searchable Symmetric Encryption (SSE), B2B Data Security, DPDP Act 2025, Privacy-Enhancing Technologies (PETs).
Abstract
A Multi-Agent Retrieval-Augmented Generation Framework for Context-Aware Legal Document Analysis
Dr. C N Shariff, Aaftab Zohra, K Sowmya, K Rakshitha, Aishwarya G
DOI: 10.17148/IJARCCE.2026.155143
Abstract: Legal document analysis requires high accuracy, traceability, and semantic understanding. While large language models (LLMs) provide strong generative capabilities, they suffer from hallucinations and lack of grounding in authoritative sources. This paper presents a Multi-Agent Retrieval-Augmented Generation (RAG) framework for legal document analysis. The system integrates semantic retrieval, vector embeddings, and collaborative agent-based reasoning to produce context-aware legal responses. A modular architecture consisting of retrieval, summarization, precedent discovery, and fact-checking agents aims to improve reliability and explainability. The framework is designed for scalable enterprise deployment and evaluated using grounding-based and qualitative evaluation metrics.
Keywords: Retrieval-Augmented Generation, Legal NLP, Multi-Agent Systems, Vector Databases
Keywords: Retrieval-Augmented Generation, Legal NLP, Multi-Agent Systems, Vector Databases
Abstract
Microstructural and Mechanical Characterization of Al2031–TiC Composites
Madevappa Channabasappa Koppal, Binit Kumar, Naresh DJ, P Ravi Kumar, Dr Santosh Janamatti, Venkatesh KC
DOI: 10.17148/IJARCCE.2026.155144
Abstract: Aluminium matrix composites (AMCs) are extensively used in aerospace, automobile, and structural engineering applications because of their high strength-to-weight ratio, excellent wear resistance, and superior mechanical performance. In the present investigation, Al2031 alloy reinforced with Titanium Carbide (TiC) particles was fabricated using the stir casting method to study the influence of reinforcement on microstructural and mechanical properties. Different weight percentages of TiC particles were incorporated into the Al2031 matrix to improve hardness, tensile strength, and wear resistance. The prepared composites were characterized using optical microscopy and scanning electron microscopy to evaluate particle distribution and interfacial bonding. Mechanical characterization was carried out through hardness and tensile testing. The results revealed that the addition of TiC particles significantly enhanced the mechanical properties of the composite due to effective load transfer and grain refinement mechanisms. Uniform distribution of reinforcement particles and reduced porosity contributed to improved performance. The developed Al2031–TiC composites exhibited promising characteristics for lightweight engineering applications.
Keywords: Aluminium Matrix Composite, Al2031 Alloy, Titanium Carbide, Stir Casting, Mechanical haracterization, Microstructure.
Keywords: Aluminium Matrix Composite, Al2031 Alloy, Titanium Carbide, Stir Casting, Mechanical haracterization, Microstructure.
Abstract
IoT-Based Smart E - Voting System using Face Authentication
Pro. Dr. R. K. Moje, Alure Omprakash Bhagwan, Surnar Amol Nagnath, Kamble Dattatray Bharat
DOI: 10.17148/IJARCCE.2026.155145
Abstract: Traditional voting processes often grapple with critical vulnerabilities. To address these security and accessibility challenges, this paper presents the design and implementation of an "IoT-Based Smart E-Voting System using Face Authentication." The proposed system leverages Internet of Things (IoT) architecture integrated with advanced biometric facial recognition to create a secure, transparent, and highly efficient electoral platform. Utilizing a Raspberry pi interfaced with a camera module, the system authenticates voters in real-time by cross-referencing live facial capture against a pre-registered, secure database. Upon successful verification, the system grants the user access to a digital ballot. The cast vote is subsequently encrypted and transmitted via an IoT network to a centralized cloud server for real-time tallying, completely eliminating the possibility of duplicate voting or unauthorized access. The integration of biometric facial authentication ensures non-repudiation, while the IoT framework provides a scalable, rapid-response infrastructure for data management. System testing demonstrates high accuracy in face detection under varied lighting conditions and minimal latency in voter registration. Ultimately, this smart e-voting framework offers a robust, cost- effective, and user-friendly alternative to conventional paper-based methods, significantly enhancing the integrity and modernization of the electoral process.
Electronic voting system evolving rapidly. This project proposes a Smart Biometric E-Voting System that integrates face recognition, fingerprint verification, and OTP (One-Time Password) authentication to achieve a multi-layered secure voting mechanism.
During user registration, the voter’s face image is captured and trained using the Haar cascade classifier, and a fingerprint sample is enrolled into the system’s database. The trained model is then stored for future authentication. At the time of voting, the voter logs into the system where facial recognition is performed using the pre-trained model. If the face is successfully authenticated, the system proceeds to fingerprint verification to confirm the voter’s identity. Additionally, an OTP is sent to the registered mobile number for final authentication. Only upon successful verification of all three credentials does the system grant access to the voting panel, allowing the voter to cast their vote.
Keywords: Raspberry Pi, Face Recognition, Fingerprint Verification, OTP Authentication, Haar cascade Classifier, Secure E-Voting System, Biometric Authentication, Machine Learning, Python, Digital Voting, IOT Etc.
Electronic voting system evolving rapidly. This project proposes a Smart Biometric E-Voting System that integrates face recognition, fingerprint verification, and OTP (One-Time Password) authentication to achieve a multi-layered secure voting mechanism.
During user registration, the voter’s face image is captured and trained using the Haar cascade classifier, and a fingerprint sample is enrolled into the system’s database. The trained model is then stored for future authentication. At the time of voting, the voter logs into the system where facial recognition is performed using the pre-trained model. If the face is successfully authenticated, the system proceeds to fingerprint verification to confirm the voter’s identity. Additionally, an OTP is sent to the registered mobile number for final authentication. Only upon successful verification of all three credentials does the system grant access to the voting panel, allowing the voter to cast their vote.
Keywords: Raspberry Pi, Face Recognition, Fingerprint Verification, OTP Authentication, Haar cascade Classifier, Secure E-Voting System, Biometric Authentication, Machine Learning, Python, Digital Voting, IOT Etc.
Abstract
Comparative Study of Air-Ground Robotic Systems and Autonomous Multi-Modal Navigation Techniques
Dr Chanda V. Reddy, Maanya Arun, Khushi K, Varsha Ranganatha, Hamsa S
DOI: 10.17148/IJARCCE.2026.155146
Abstract: Air-ground robotic systems that combine UAVs and UGVs are widely used in disaster response, firefighting, reconnaissance, inspection, and autonomous navigation. These hybrid systems combine the mobility of the air with the stability and endurance of the ground. This study provides a comparative study of fifteen research papers on hybrid robots based on navigation, localisation, locomotion, firefighting, and autonomous coordination. It includes mechanisms, control strategies, energy efficiency, and GPS-denied navigation techniques. The main trends are topology optimisation, adaptive morphogenesis, multi-sensor localisation, and behaviour-based coordination. The table compares system benefits, limitations and performance. The study concludes that the hybrid robotic systems enhance flexibility and efficiency, but there are challenges in localisation, payload optimisation, and autonomous decision-making [1]–[15].
Keywords: Air-Ground Robots, UAV-UGV Integration, Hybrid Robotics, Autonomous Navigation, Multi-Modal Locomotion, GPS-Denied Localisation, Firefighting Robots, Comparative Study, Morphing Robots, Robotics Control Systems
Keywords: Air-Ground Robots, UAV-UGV Integration, Hybrid Robotics, Autonomous Navigation, Multi-Modal Locomotion, GPS-Denied Localisation, Firefighting Robots, Comparative Study, Morphing Robots, Robotics Control Systems
Abstract
Real -Time Medicine Tracking and Distribution Coordination System
Ms. A. Nisha, ME., Guruprasath M, Nagoor Meeran T
DOI: 10.17148/IJARCCE.2026.155147
Abstract: The project not only addresses privacy and security needs but also the issue of trust in data sharing practices. By providing a privacy-preserving system that guarantees the confidentiality of individuals' information, the project aims at creating a secure and reliable environment for data sharing. Additionally, the proposed system will support the implementation of various use cases such as health care, banking, and e-commerce where data privacy and security are of paramount importance. The integration of various advanced techniques like decision tree, random forest, or support vector machine for data sensitivity assessment and the selection of the most suitable cryptographic technique for the different types of data to be processed will be also feasible. This will help in providing the best possible solution in terms of data privacy and accessibility.Ultimately, the project is expected to make a strong contribution to the development of privacy-preserving technologies through its innovative approach and the development of the proposed Intelligent Privacy Preserving Data Encryption and Anonymization System. In the long run, the project's results might have a significant impact on the future of data protection technologies as more and more organizations migrate to cloud services and share information across borders.
Keywords: Data protection, encryption, anonymization, privacy-preserving, cloud computing, big data analytics, interconnected systems, decision making, automated systems, trust
Keywords: Data protection, encryption, anonymization, privacy-preserving, cloud computing, big data analytics, interconnected systems, decision making, automated systems, trust
Abstract
Carbon Footprint Tracking in Logistics System: A Survey of AI-Based Emission Monitoring Approaches
Bhagyashri K Kulkarni, Shashank Urs H P, Vinod, Sandeep T Rathod, and Mayur Kishan Rathod
DOI: 10.17148/IJARCCE.2026.155148
Abstract: The logistics and transportation industry is one of the leading contributors to global carbon dioxide (CO₂) emissions. With the rapid growth of e-commerce and supply chain operations, accurately monitoring and reducing carbon footprints in logistics has become a critical environmental and operational challenge. This paper presents a survey of existing research on AI-driven carbon emission monitoring, route optimization, green logistics, and machine learning- based prediction systems. Six key studies are analyzed and compared across dimensions such as methodology, dataset, emission parameters, AI models, and limitations. Based on this survey, we propose a comprehensive web-based Carbon Footprint Tracking System for logistics operations that integrates machine learning-based emission prediction, AI- powered route optimization, shipment-wise carbon tracking, and a centralized analytics dashboard. The proposed system aims to address key gaps in existing approaches by combining real-time tracking, future emission forecasting, and actionable carbon reduction recommendations in a unified platform.
Keywords: Carbon Footprint, Logistics Emissions, Green Logistics, Route Optimization, Machine Learning, CO₂ Prediction, Sustainable Transportation, Emission Tracking, Artificial Intelligence, Smart Logistics
Keywords: Carbon Footprint, Logistics Emissions, Green Logistics, Route Optimization, Machine Learning, CO₂ Prediction, Sustainable Transportation, Emission Tracking, Artificial Intelligence, Smart Logistics
Abstract
KG-CTCN: Knowledge-Guided Causal Temporal Convolutional Networks for Event-Driven Sugarcane Red Rot Forecasting
Utkarsh Joshilkar, Prasanna Bandiwadekar, Aditya Chatte, Kousen Attar, Kshitij Aiwale
DOI: 10.17148/IJARCCE.2026.155149
Abstract: The prevention of Sugarcane Red Rot (Colletotrichum falcatum) outbreaks remains a primary challenge in tropical agriculture, where traditional predictive models often suffer from “monsoon bias”—mistaking seasonal humidity for specific disease triggers. This paper presents KG-CTCN, a Knowledge-Guided Causal Temporal Convolutional Network designed as an event-driven early warning system. Moving beyond daily binary classification, KG-CTCN models 28-day environmental trajectories and integrates agronomic constraints via a cross-modal attention fusion layer. We reconstruct the system's development process, from initial baseline failures compromised by temporal feature leakage to the current deployment-ready architecture. Experimental results on a historical validation set (2019– 2021) demonstrate that KG-CTCN achieves 80% detection of recorded major outbreak events with an average lead time of 12.5 days and a false positive rate of 15.2%—a deliberate design tradeoff reflecting the asymmetric cost structure of outbreak forecasting, where missed detections incur catastrophic crop loss while false advisories impose only marginal spray costs. Robustness tests involving synthetic temporal shifts confirm that the model relies on physical causal signals rather than chronological memorization, marking a significant step toward trustworthy AI in plant pathology.
Keywords: Crop Disease Prediction, Sugarcane Red Rot, Temporal Convolutional Networks, Knowledge-Guided Machine Learning, Early Warning Systems.
Keywords: Crop Disease Prediction, Sugarcane Red Rot, Temporal Convolutional Networks, Knowledge-Guided Machine Learning, Early Warning Systems.
Abstract
Exam Section Automation for Higher Educational Institutions
Hayath T M, Yashitha C, Vidyashree, Y Indusree, Yashodha
DOI: 10.17148/IJARCCE.2026.155150
Abstract: The administration of examinations in academic institutions continues to rely predominantly on manual, paper- based workflows, resulting in operational delays, inconsistencies, and a heightened likelihood of human error. Critical tasks such as hall ticket preparation, seating arrangement allocation, and dissemination of exam-related information demand substantial administrative effort and often fail to ensure timely and accurate communication to students. Existing digital solutions offer partial automation but lack robustness, scalability, and secure mechanisms for handling examination-sensitive data. This study presents the Exam Section Automation System, an integrated and automated platform designed to streamline the pre-examination workflow through algorithmic hall ticket generation, systematic seat allocation, and automated email-based notifications.
Keywords: Examination Management, Academic Automation, Hall Ticket Generation, Seating Allocation, Secure Notification Systems.
Keywords: Examination Management, Academic Automation, Hall Ticket Generation, Seating Allocation, Secure Notification Systems.
Abstract
SKY SHIELD A SKETCH-BASED DEFENSE SYSTEM AGAINST APPLICATION LAYER DDOS ATTACK
Mrs. R. Janaki, ME., Srikanth S, Krishnaraj K
DOI: 10.17148/IJARCCE.2026.155151
Abstract: The rapid growth of digital technologies, cloud computing, online learning platforms, and decentralized communication systems has increased the demand for secure academic credential management and intelligent cybersecurity solutions. Traditional credential verification systems mainly rely on centralized databases, which are vulnerable to data tampering, certificate forgery, unauthorized access, privacy breaches, and cyber attacks. To overcome these limitations, this paper proposes an AI-powered Blockchain Academic Validation Engine (BAVE-Chain) integrated with decentralized storage and intelligent cyber attack detection mechanisms.
The proposed system combines blockchain technology, smart contracts, InterPlanetary File System (IPFS), machine learning algorithms, and real-time monitoring techniques to provide secure credential verification and advanced cybersecurity protection. Blockchain technology ensures tamper-resistant storage of academic records and verification transactions, while IPFS provides decentralized and secure storage of certificates and supporting documents. Smart contracts automate verification and approval processes, improving transparency and reducing manual intervention.
The system also integrates machine learning algorithms such as Random Forest, Support Vector Machine (SVM), and XGBoost for analyzing network traffic patterns and detecting malicious activities in real time. The proposed framework provides functionalities including student registration, certificate generation, decentralized credential storage, blockchain verification, traffic monitoring, alert generation, and dashboard visualization through Streamlit-based interfaces.
Experimental analysis and system testing demonstrate that the proposed framework effectively improves credential security, verification transparency, decentralized data management, and cyber attack detection accuracy. The integration of blockchain technology, artificial intelligence, and decentralized storage significantly reduces the risks of credential forgery, unauthorized access, and malicious network activities. Overall, the proposed BAVE-Chain framework offers a secure, scalable, intelligent, and reliable solution for modern academic credential validation and cybersecurity applications.
Keywords: Blockchain, Academic Credential Verification, BAVE-Chain, Smart Contracts, IPFS, Machine Learning, Cyber Attack Detection, Artificial Intelligence, Decentralized Storage, Network Security, Credential Validation, Random Forest, SVM, XGBoost, Cybersecurity.
The proposed system combines blockchain technology, smart contracts, InterPlanetary File System (IPFS), machine learning algorithms, and real-time monitoring techniques to provide secure credential verification and advanced cybersecurity protection. Blockchain technology ensures tamper-resistant storage of academic records and verification transactions, while IPFS provides decentralized and secure storage of certificates and supporting documents. Smart contracts automate verification and approval processes, improving transparency and reducing manual intervention.
The system also integrates machine learning algorithms such as Random Forest, Support Vector Machine (SVM), and XGBoost for analyzing network traffic patterns and detecting malicious activities in real time. The proposed framework provides functionalities including student registration, certificate generation, decentralized credential storage, blockchain verification, traffic monitoring, alert generation, and dashboard visualization through Streamlit-based interfaces.
Experimental analysis and system testing demonstrate that the proposed framework effectively improves credential security, verification transparency, decentralized data management, and cyber attack detection accuracy. The integration of blockchain technology, artificial intelligence, and decentralized storage significantly reduces the risks of credential forgery, unauthorized access, and malicious network activities. Overall, the proposed BAVE-Chain framework offers a secure, scalable, intelligent, and reliable solution for modern academic credential validation and cybersecurity applications.
Keywords: Blockchain, Academic Credential Verification, BAVE-Chain, Smart Contracts, IPFS, Machine Learning, Cyber Attack Detection, Artificial Intelligence, Decentralized Storage, Network Security, Credential Validation, Random Forest, SVM, XGBoost, Cybersecurity.
Abstract
Optimised Battery Thermal Management System: Design and Thermal Analysis Review
Syed Mohammd Ashfaque, C Mohhammad Asif, Shaik Naveed, Altaf
DOI: 10.17148/IJARCCE.2026.155152
The rapid transition toward electric
mobility is driven by the urgent need to reduce greenhouse gas emissions and
dependency on fossil fuels [1]. At the core of this transition are Lithium-ion
batteries, favored for their high energy density and specific power [1].
However, these batteries are highly sensitive to temperature; their optimal
operating range is typically between 15°C and 35°C [1]. Operating outside this
range can lead to accelerated degradation, reduced range, and catastrophic
thermal runaway events [2]. Consequently, a robust Battery Thermal Management
System is essential to maintain temperature uniformity and prevent excessive
heat buildup during high charge-discharge cycles [3]. This study focuses on
designing and analyzing an optimized BTMS to enhance cooling efficiency and
safety for modern electric vehicles.
Abstract
“Recipe Generation from Food Image Using Deep Learning”: A Comprehensive Review
Vijaya Durga H, Sree vishnu, Dadavali H
DOI: 10.17148/IJARCCE.2026.155153
Abstract: Job Food recognition and recipe generation have become important applications of Artificial Intelligence and Deep Learning in modern food technology. This paper presents “Recipe Generation from Food Image Using Deep Learning,” an intelligent system that automatically identifies food items from uploaded images and generates recipe titles, ingredients, and cooking instructions. The system uses Convolutional Neural Networks (CNNs) for food image feature extraction and Transformer-based Encoder-Decoder architectures with attention mechanisms for recipe generation. Natural Language Processing (NLP) techniques are applied to process ingredient lists and cooking instructions efficiently. The application is developed using Python, Flask, TensorFlow, and PyTorch, along with a web-based interface for realtime user interaction. The proposed system reduces manual recipe searching effort, improves user convenience, and provides intelligent cooking assistance. It can be applied in smart cooking assistants, restaurant systems, food recommendation platforms, and AI-based kitchen automation systems.
Keywords: Recipe Generation, Food Image Recognition, Deep Learning, CNN
Keywords: Recipe Generation, Food Image Recognition, Deep Learning, CNN
Abstract
A Modular Data Deduplication Framework with Support for Multi-Format Document Analysis
Shreya Modi, Sreenivasa M, Shridevi, Shreya Y.P, Trisha Prameela Y
DOI: 10.17148/IJARCCE.2026.155154
Abstract: Data deduplication is a critical technique used to reduce repeated storage of documents in digital systems. In many organizational and academic environments, identical files or portions of files are stored multiple times, leading to increased storage consumption and management complexity. This research presents a comprehensive document-based deduplication framework that provides native support for PDF, DOCX, and plain text file formats. The proposed system processes uploaded documents by extracting textual content and systematically dividing it into smaller, manageable chunks. A cryptographic hash-based comparison methodology is employed to determine whether content segments already exist within the storage repository. When duplicate content is identified, the system maintains reference pointers rather than storing redundant copies, thereby achieving significant storage optimization. Experimental evaluation demonstrates that the chunk-level approach successfully identifies partial duplicates that would be missed by traditional file-level comparison methods. The framework is designed for practical deployment in academic institutions and organizational document management systems where efficient handling of duplicate content is essential.
Keywords: Data deduplication, document analysis, hashbased comparison, chunk-level processing, storage optimization, multi-format support.
Keywords: Data deduplication, document analysis, hashbased comparison, chunk-level processing, storage optimization, multi-format support.
Abstract
A Survey on Machine Learning and Deep Learning Techniques for Fake Review Detection
Ms. Samruddhi P. Ingale, Dr. V. H. Deshmukh, Dr. P. P. Deshmukh
DOI: 10.17148/IJARCCE.2026.155155
Abstract: Online reviews have become an essential source of information for consumers when evaluating products and services on digital platforms. However, the growing presence of deceptive or fake reviews has raised concerns about the reliability of these systems. Over the past decade, researchers have proposed various techniques to detect such reviews using natural language processing, machine learning, and deep learning methods. This paper presents a comprehensive survey of existing approaches for fake review detection, focusing on different categories of techniques including traditional machine learning models, neural network-based methods, and hybrid approaches that combine textual and behavioral features. The survey examines commonly used feature extraction techniques such as term frequency–inverse document frequency, sentiment analysis, and reviewer behavior analysis. It also discusses recent developments in deep learning models that capture contextual relationships within review text. In addition, the paper highlights key challenges faced in this domain, including limited availability of labeled datasets, data imbalance, and evolving strategies used by spammers. Finally, the study identifies important research gaps and suggests directions for future work, particularly in improving model interpretability and developing real-time detection systems. The findings provide a structured understanding of existing methods and their limitations, which can support further research in this area.
Keywords: Fake review detection, opinion spam, machine learning, deep learning, sentiment analysis, text mining, behavioral analysis, feature engineering, survey, review classification.
Keywords: Fake review detection, opinion spam, machine learning, deep learning, sentiment analysis, text mining, behavioral analysis, feature engineering, survey, review classification.
Abstract
AI-Powered Deepfake Detection and Liveness Detection
Dr. Jagadish R M, Chetana HK, K Shashikala, Mounika M, Pushpitha JR
DOI: 10.17148/IJARCCE.2026.155156
Abstract: Advances in deep learning have made it easier to create realistic visual content, popularly known as deepfakes. While such tools find applications in creative media, the serious risks of identity theft, misinformation, and digital manipulation also come with them. This work presents a lightweight detection framework that identifies manipulated visual content and verifies whether the source is a real live person. The system proposed here is empowered with both MobileNet and custom CNN models to analyze facial behavior, expression dynamics, and minute texture variations that distinguish genuine recordings from spoofing attempts created using printed images, masks, or replayed clips. For real- time processing of both images and videos, a web-based interface is developed using Flask. Experimental evaluations demonstrate accuracy close to 90%, thus extending the applicability of the proposed solution to secure authentication environments and digital forensics.
Keywords: Deepfake Detection, Liveness Detection, MobileNet, CNN, AI, Flask Web Application.
Keywords: Deepfake Detection, Liveness Detection, MobileNet, CNN, AI, Flask Web Application.
Abstract
Edu Site: A Advanced Learning Platform Using AI-powered Education and Personalized Learning
Kiran Mudaraddi∗, Prem Kumar B T, Prerana B, and R M Sneha
DOI: 10.17148/IJARCCE.2026.155157
Abstract: The Advanced Learning Platform (ALP) is a web-based educational system that uses modern AI services to provide role-based, adaptive instructional content and visual aids. ALP integrates large language models for text generation and diffusion-based models for educational image creation to support students, teachers, professors, and researchers with tailored ex-planations, diagrams, and learning artifacts. The implementation uses a Node.js + Express backend, MongoDB for persistent storage, and a responsive frontend. Key contributions are: role-aware content generation and formatting, automated educational visual synthesis, and audit-ready activity logging to support learning analytics and reproducibility. We present the system architecture, API design, security and scalability considerations, an evaluation plan for accuracy/usability, and discuss ethical safeguards and limitations.
Keywords: Artificial Intelligence in Education (AIED), Adaptive Learning Systems, Intelligent Tutoring Systems (ITS), Large Language Models (LLMs), Educational Content Gen-eration, AI-Driven Personalization, Role-Based Learning Plat-forms, Multimodal Learning Technologies, Generative Artificial Intelligence, Educational Image Generation, Learning Analyt-ics, Web-Based Educational Platforms, Node.js and MongoDB Architecture, Human-AI Collaboration in Education, Scalable Educational Technology
Keywords: Artificial Intelligence in Education (AIED), Adaptive Learning Systems, Intelligent Tutoring Systems (ITS), Large Language Models (LLMs), Educational Content Gen-eration, AI-Driven Personalization, Role-Based Learning Plat-forms, Multimodal Learning Technologies, Generative Artificial Intelligence, Educational Image Generation, Learning Analyt-ics, Web-Based Educational Platforms, Node.js and MongoDB Architecture, Human-AI Collaboration in Education, Scalable Educational Technology
Abstract
A NOVEL BRAIN-COMPUTER INTERFACE FRAMEWORK FOR REAL-TIME CONTROL AND COMMUNICATION USING NEURAL SIGNAL PROCESSING
Sara Vaivswat, M Sri Durgaa, Simran Shiva Prasad T, R K Nithish, Pranav Sankar, R Tarun Gorani, Dr Sonia Maria D’Souza
DOI: 10.17148/IJARCCE.2026.155158
Abstract: Brain-Computer Interface (BCI) technology represents a transformative advancement in neuroengineering by establishing direct communication pathways between the human brain and external devices. The primary objective of BCI systems is to acquire, process, and interpret neural signals to restore, enhance, or supplement cognitive and motor functions, particularly for individuals affected by paralysis, neurological disorders, or communication impairments. Conventional BCI systems often face significant limitations such as signal attenuation due to skull interference, thrombosis risks, thermal damage, limited long-term biocompatibility, and power inefficiencies. This paper proposes an advanced multi-stentrode adaptive vascular BCI architecture designed to address these challenges through minimally invasive intravascular implantation, enabling improved neural signal acquisition without skull obstruction. The system integrates multiple vascular stentrodes for enhanced signal accuracy, ultrasound-based wireless communication for safe data transmission, ASIC-based low-power intelligent processing for thermal regulation, and dual-layer protective coatings including heparin, PEG, graphene, and titanium for clot prevention, biocompatibility, and heat dissipation. Additionally, the framework incorporates shape-adaptive materials, dynamic drug release systems, embedded biosensors for clot and pressure monitoring, and hybrid self-powering mechanisms utilizing biochemical, piezoelectric, and ultrasound energy sources. Through the integration of adaptive biomaterials, intelligent safety systems, and real-time physiological monitoring, this design significantly improves long-term implant safety, vascular compatibility, and operational sustainability. The proposed system demonstrates the potential to revolutionize next-generation BCI applications in assistive healthcare, neuroprosthetics, rehabilitation, and advanced medical monitoring, paving the way for safer, more efficient, and clinically viable brain-computer interface technologies.
Keywords: Brain-Computer Interface (BCI), Stentrode, Neuroengineering, Minimally Invasive Neural Interface, Wireless Neural Interface, Real-Time Monitoring.
Keywords: Brain-Computer Interface (BCI), Stentrode, Neuroengineering, Minimally Invasive Neural Interface, Wireless Neural Interface, Real-Time Monitoring.
Abstract
Hybrid Renewable Energy System for Medical Supply Chains
M.Lokesh, Sharanappa.Y, Kalyan Kumar.A, Karegowda.A, Jayaprakash.B
DOI: 10.17148/IJARCCE.2026.155159
Abstract: Medical supply chains play a critical role in ensuring the availability of medicines, vaccines, blood samples, and emergency healthcare equipment. Continuous power supply is essential for maintaining cold storage systems, transportation monitoring, and healthcare logistics operations. Conventional energy sources are associated with high operational costs, fuel dependency, and environmental pollution. To overcome these limitations, this paper proposes a Hybrid Renewable Energy System (HRES) for medical supply chains integrating solar energy, wind energy, battery storage, and IoT-based monitoring.
The proposed system aims to provide uninterrupted and sustainable energy for medical storage units, vaccine refrigeration systems, and transportation hubs. Solar photovoltaic panels and wind turbines are used as primary renewable energy sources, while battery storage ensures continuous operation during power fluctuations or low renewable energy availability. IoT sensors are incorporated for real-time monitoring of temperature, humidity, battery condition, and energy consumption.
The proposed hybrid system improves energy efficiency, reduces operational costs, minimizes carbon emissions, and enhances the reliability of medical supply chain infrastructure. The system is particularly suitable for rural and remote healthcare centers where grid power availability is limited.
Keywords: Hybrid Renewable Energy System, Medical Supply Chain, Solar Energy, Wind Energy, IoT, Battery Storage, Sustainable Healthcare.
The proposed system aims to provide uninterrupted and sustainable energy for medical storage units, vaccine refrigeration systems, and transportation hubs. Solar photovoltaic panels and wind turbines are used as primary renewable energy sources, while battery storage ensures continuous operation during power fluctuations or low renewable energy availability. IoT sensors are incorporated for real-time monitoring of temperature, humidity, battery condition, and energy consumption.
The proposed hybrid system improves energy efficiency, reduces operational costs, minimizes carbon emissions, and enhances the reliability of medical supply chain infrastructure. The system is particularly suitable for rural and remote healthcare centers where grid power availability is limited.
Keywords: Hybrid Renewable Energy System, Medical Supply Chain, Solar Energy, Wind Energy, IoT, Battery Storage, Sustainable Healthcare.
Abstract
AI-BASED RESUME ANALYZER
Mrs. Sameena Yasmeen, N Indhu, Nancy Mittal, P Harika, P Saima
DOI: 10.17148/IJARCCE.2026.155160
Abstract: The need for this project is to develop an AI-based resume analyzer system that automates and optimizes the resume screening and candidate evaluation process. The proposed system utilizes artificial intelligence (AI) and natural language processing (NLP) techniques to extract and analyze important information from resumes, such as personal details, educational qualifications, technical skills, soft skills, certifications, and work experience. Indeed, the system is capable of comparing the extracted resume content with predefined job descriptions to identify the most suitable candidates based on skill matching and relevance. This system is expected to reduce the time and effort involved in manual resume screening while minimizing human bias and improving recruitment accuracy. The proposed system involves intelligent text processing, resume classification, keyword extraction, and skill gap analysis to provide accurate recommendations for both recruiters and jobseekers. Moreover, the system has the capability of learning from resume patterns and job requirements, thereby improving its efficiency and decision-making ability over time. The result of this project is expected to be a fast, reliable, and accurate resume analysis system that assists recruiters in shortlisting candidates effectively and helps applicants improve their resumes.
Keywords: Artificial Intelligence, Resume Analyzer, Natural Language Processing, Resume Screening, Skill Matching, Recruitment Automation.
Keywords: Artificial Intelligence, Resume Analyzer, Natural Language Processing, Resume Screening, Skill Matching, Recruitment Automation.
Abstract
Semantic And Sentiment Analysis of Multilingual Code-Mixed Text
Saritha D, Bhoomika LM, Lavanya Shetty, Raksha K, Reeta Gracy
DOI: 10.17148/IJARCCE.2026.155161
Abstract: In multilingual societies, communication frequently involves the mixing of two or more languages within a single sentence. This phenomenon, known as code-mixing, is commonly observed in everyday conversations, social media interactions, and spoken communication. Languages such as English, Hindi, and Kannada are often combined, creating challenges for traditional Natural Language Processing (NLP) systems that are primarily designed for monolingual data. The objective of this project is to develop a system capable of performing semantic interpretation and sentiment analysis of multilingual code-mixed text. The proposed system will initially process text-based inputs containing mixed language content and generate a normalized English representation of the input sentence. In addition, the system will classify the sentiment of the input as positive, negative, or neutral and provide a confidence score for the predicted sentiment. The project leverages pre-trained multilingual transformer models and language processing frameworks to analyse mixed-language inputs effectively. As an extension, the system may also explore speech input processing, where spoken sentences are converted to text using speech recognition techniques before being analysed. The proposed system demonstrates how modern NLP techniques can be applied to understand multilingual communication and improve the processing of mixed-language textual data.
Keywords: Multilingual Code-Mixing, Sentiment Analysis, Natural Language Processing (NLP), Semantic Interpretation, Multilingual Transformer Models, Text Normalization.
Keywords: Multilingual Code-Mixing, Sentiment Analysis, Natural Language Processing (NLP), Semantic Interpretation, Multilingual Transformer Models, Text Normalization.
Abstract
STUDY ON PROPETIES BANNANA FIBER COMPOSITIONS
Lokesh P, Manoj D P, Honnurswamy P, Srinidhi N P
DOI: 10.17148/IJARCCE.2026.155162
Abstract: This study investigates the chemical, physical, and mechanical properties of banana fiber extracted from the pseudostem of the Musa sapientum plant as a sustainable alternative to synthetic reinforcements. Banana fibers are primarily composed of cellulose (47.1%–63.4%), hemicellulose (15%–27.8%), and lignin (9.8%–27.7%) [1], [2], [3]. While these fibers demonstrate high specific strength, their inherent hydrophilicity often results in poor interfacial bonding with polymer matrices [3], [4]. Experimental results indicate that raw fibers exhibit a crystallinity index of approximately 52% and a density range of 0.22 g/cm³ to 1.03 g/cm³ [2], [5]. Mechanical characterization reveals tensile strengths varying from 69.99 MPa to 743.9 MPa, depending on cultivation conditions and extraction methods [1], [2], [6]. To address performance limitations, this research highlights the impact of surface modifications, specifically 5% NaOH alkali treatment, which has been shown to enhance tensile strength by up to 52% and reduce moisture absorption by 39% by removing non-cellulosic impurities [7]. The findings suggest that with optimized chemical treatments, banana fiber is a viable, eco-friendly resource for high-performance applications in the automotive and textile industries [1], [8].
Keywords: Banana fiber; Musa sapientum; Cellulose; Alkali treatment; Mechanical properties; Natural fiber composites; Sustainability.
Keywords: Banana fiber; Musa sapientum; Cellulose; Alkali treatment; Mechanical properties; Natural fiber composites; Sustainability.
Abstract
“SOLAR REFRIGERATION SYSYTEM USING PELTIER MODULES”
Keshava Murthy U, Lokesh B, Manjunath B, Manjunatha K M
DOI: 10.17148/IJARCCE.2026.155163
Abstract: Traditional vapor compression refrigeration systems account for approximately 15% of global electrical energy demand and rely on refrigerants that contribute significantly to ozone depletion and global warming [1], [2]. This research investigates the design and performance of a sustainable alternative: a solar-powered thermoelectric refrigerator utilizing the Peltier effect. The system integrates 12V DC thermoelectric modules with a photovoltaic array and battery storage, eliminating the need for moving parts or chemical refrigerants [3], [4]. Experimental results demonstrate that optimized systems can achieve a Coefficient of Performance as high as 3.74 at a current of 1.4A when the cold side temperature reaches 0°C [5]. Performance tests on a 2.6-liter cooling chamber showed a rapid temperature reduction from an ambient 22.48°C to 15.15°C within 136 seconds [3]. Furthermore, the integration of dual-axis solar tracking can enhance the COP by 44% to 75% compared to fixed-panel systems [6]. The results highlight that while thermoelectric coolers typically have lower COPs than conventional systems, they offer a 44–63% reduction in power consumption for small-scale applications and a potential lifespan of up to 40 years [3], [7]. These characteristics make solar-powered Peltier refrigeration an ideal solution for portable medical storage and food preservation in remote, off-grid, and humanitarian contexts [8], [9].
Keywords: Peltier Effect, Thermoelectric Cooling, Solar Photovoltaic, Coefficient of Performance, Sustainable Refrigeration, Portable Cooling, Renewable Energy, Heat Dissipation
Keywords: Peltier Effect, Thermoelectric Cooling, Solar Photovoltaic, Coefficient of Performance, Sustainable Refrigeration, Portable Cooling, Renewable Energy, Heat Dissipation
Abstract
“Design, Development, and Analysis of a Solar Thermal Powered Automatic Air Purification System’’
Teja Raju P, Samil Raju P, Tharun Varma Raju V, Nabirashul P
DOI: 10.17148/IJARCCE.2026.155164
Abstract: Air pollution in urban and industrial regions remains a significant threat to public health, often exacerbated by the high energy consumption and carbon emissions of conventional grid-powered purification systems (Sharma et al., 2025; Sharyani et al., 2025). This study presents the design and analysis of an automatic air purification system powered by solar thermal and photovoltaic energy (Sharyani et al., 2025; Suri et al., 2025). The proposed system utilizes a multi- stage filtration process, incorporating activated carbon and HEPA filters to effectively eliminate airborne contaminants (Sharma et al., 2025; Suri et al., 2025).
Field tests of the solar-powered prototype demonstrate an average removal efficiency of 90% for particulate matter (PM2.5) and a substantial reduction in indoor volatile organic compounds (Sharma et al., 2025). The integration of intelligent sensor technology and microcontrollers allows the system to monitor air quality in real-time and adjust its purification settings dynamically to optimize energy use (Sharyani et al., 2025; Suri et al., 2025). Furthermore, performance evaluations of hybrid solar collectors indicate that a total efficiency of 44.68% to 58.59% is achievable under varied conditions (Thinsurat et al., 2018). This research highlights a sustainable, low-maintenance solution for air quality management in environments such as dental clinics and healthcare facilities, where maintaining sterile air is critical (Sharyani et al., 2025).
Keywords
• Solar Thermal Energy • Air Purification • HEPA Filtration • PM2.5 Removal Efficiency • Photovoltaic-Thermal • Automated Air Quality Monitoring • Sustainable Healthcare • Volatile Organic Compounds • Renewable Energy Systems • Urban Air Remediation
Field tests of the solar-powered prototype demonstrate an average removal efficiency of 90% for particulate matter (PM2.5) and a substantial reduction in indoor volatile organic compounds (Sharma et al., 2025). The integration of intelligent sensor technology and microcontrollers allows the system to monitor air quality in real-time and adjust its purification settings dynamically to optimize energy use (Sharyani et al., 2025; Suri et al., 2025). Furthermore, performance evaluations of hybrid solar collectors indicate that a total efficiency of 44.68% to 58.59% is achievable under varied conditions (Thinsurat et al., 2018). This research highlights a sustainable, low-maintenance solution for air quality management in environments such as dental clinics and healthcare facilities, where maintaining sterile air is critical (Sharyani et al., 2025).
Keywords
• Solar Thermal Energy • Air Purification • HEPA Filtration • PM2.5 Removal Efficiency • Photovoltaic-Thermal • Automated Air Quality Monitoring • Sustainable Healthcare • Volatile Organic Compounds • Renewable Energy Systems • Urban Air Remediation
Abstract
Hybrid Embedding Model for Document Classification
Ranjana S. Chakrasali, Chandana A. Athreyesa, K. Shridevi B. Adiga, Vathsala, Vishnupriya
DOI: 10.17148/IJARCCE.2026.155165
Abstract: Managing large collections of digital documents has become increasingly difficult in academic and professional environments. Files such as research papers, reports, PDFs, and project documents are often stored without proper organization, making retrieval slow and inefficient. This work proposes a hybrid document classification framework that combines TF-IDF statistical features with contextual embeddings generated using BERT. The combined representation helps the model capture both important keywords and semantic meaning from documents. A lightweight classification layer is used to assign uploaded files into categories such as Business, Politics, Sports, Health, and Technology. In addition, a rule-based file extension classifier is integrated to improve efficiency for commonly identifiable file types. A Flask-based web interface enables users to upload documents and automatically organize them into category folders. Experimental evaluation on the BBC News dataset demonstrates that the proposed hybrid model performs better than standalone TF-IDF and BERT models in terms of classification accuracy and Macro F1-score.
Keywords: document classification, hybrid embedding, TF-IDF, BERT, natural language processing, feature fusion, Flask, text categorization
Keywords: document classification, hybrid embedding, TF-IDF, BERT, natural language processing, feature fusion, Flask, text categorization
Abstract
DESIGN OF AUTOMATION SEALPOT USED IN GASLINE TO REMOVE CONDESATE WATER
Mohammed Kaif, Mohammed Nawaz, Mohammed Gayaz, S Mohammed Tousif
DOI: 10.17148/IJARCCE.2026.155166
Abstract: The accumulation of liquid condensate and water in natural gas pipelines is a critical operational issue that leads to internal corrosion, hydrate formation, and significant reductions in energy efficiency [1], [2], [3]. Traditional manual drainage methods often result in inconsistent maintenance and accidental greenhouse gas emissions [4], [5]. This study proposes the design of an automated seal pot system specifically engineered for the continuous removal of these liquids. By integrating high-accuracy liquid level sensors with PID-based control logic and automatic solenoid valves, the system ensures real-time response to condensate accumulation [4], [6], [7]. Implementation of similar automated technologies has demonstrated a potential 13% increase in production by maintaining optimal flow regimes and preventing the formation of "ice plugs" or slugs [2], [7]. The proposed design offers a scalable and environmentally safer alternative to manual drips, contributing to the digitalization and integrity management of both high-pressure trunk lines and low-pressure gas gathering networks [4], [8].
Keywords: Natural Gas Pipelines; Condensate Removal; Automated Seal Pot; Liquid Accumulation; Pipeline Integrity; Smart Control Systems; Flow Efficiency.
Keywords: Natural Gas Pipelines; Condensate Removal; Automated Seal Pot; Liquid Accumulation; Pipeline Integrity; Smart Control Systems; Flow Efficiency.
Abstract
AI-Based Smart Mobile Robotic Arm with Adaptive Gripping System
Mohammad Ibrahim, Chandan Kumar Kushwaha, Mohammed Zuhair C, Mohammad farooq
DOI: 10.17148/IJARCCE.2026.155167
Abstract: This research presents the design and development of an AI-based smart mobile robotic arm with an adaptive gripping system, designed to address critical challenges in autonomous manipulation and intelligent automation. The proposed system integrates advanced artificial intelligence algorithms with mobile manipulation capabilities, enabling autonomous navigation in complex environments through multi-sensor fusion and real-time environmental perception. At the core of the system is a sophisticated computer vision framework employing lightweight Convolutional Neural Network models and OpenCV for precise object detection, recognition, and localization, allowing the robotic arm to identify and manipulate objects with varying shapes, sizes, and orientations without prior knowledge of their positions. The adaptive gripping mechanism dynamically adjusts to object characteristics, ensuring reliable grasping across diverse scenarios. The system architecture combines a mobile platform with a multi-degree-of-freedom robotic arm, controlled through embedded systems such as Raspberry Pi, facilitating seamless coordination between perception, motion planning, and manipulation modules. Key innovations include real-time image processing, dynamic path planning, and intelligent grasp strategies that enable the system to operate effectively in unstructured environments where object locations are not predefined. Experimental validation demonstrates the system's capability to perform autonomous pick-and-place operations with high accuracy and adaptability, making it suitable for applications in intelligent warehousing, industrial automation, material handling, and assistive robotics. The integration of AI-driven perception with robust mechanical design represents a significant advancement in autonomous robotic manipulation technology, offering a scalable and cost-effective solution for modern automation challenges.
Keywords: Artificial Intelligence, Mobile Robotic Arm, Adaptive Gripping System, Computer Vision, Object Detection, Autonomous Navigation, Pick-and-Place Operations, Multi-Sensor Fusion, Real-Time Image Processing, Intelligent Automation, Robotic Manipulation, Deep Learning, Embedded Control, Motion Planning, Human-Robot Interaction
Keywords: Artificial Intelligence, Mobile Robotic Arm, Adaptive Gripping System, Computer Vision, Object Detection, Autonomous Navigation, Pick-and-Place Operations, Multi-Sensor Fusion, Real-Time Image Processing, Intelligent Automation, Robotic Manipulation, Deep Learning, Embedded Control, Motion Planning, Human-Robot Interaction
Abstract
“Design & Performance Analysis of EOT Crane, Modification from DC Electromagnetic to Thruster Brake System”
Mohammed Muzameel, N Momin, Mohammed Kaif, C D Mounesh
DOI: 10.17148/IJARCCE.2026.155168
Abstract: The Electrical Overhead Travelling (EOT) crane is one of the most widely used material handling systems in industrial applications such as steel plants, manufacturing industries, warehouses, and construction sectors. The braking system of an EOT crane plays a vital role in ensuring operational safety, load control, reliability, and smooth functioning. Conventional DC electromagnetic brake systems used in cranes often face problems such as high maintenance, overheating, brake coil failure, and reduced braking efficiency during continuous operation. To overcome these limitations, the present study focuses on the design and performance analysis of an EOT crane by modifying the existing DC electromagnetic brake system into an electro-hydraulic thruster brake system.
The project involves studying the working principles of both braking systems and comparing their performance based on braking torque, response time, reliability, energy consumption, maintenance requirements, and operational safety. The design analysis includes load calculations, braking force estimation, motor compatibility, and stress considerations under different operating conditions. The thruster brake system provides fail-safe operation, faster response, better heat dissipation, and improved braking performance compared to the conventional DC electromagnetic brake.
The modification aims to enhance crane efficiency, reduce downtime, improve safety standards, and minimize maintenance costs. The results obtained from the analysis indicate that the electro-hydraulic thruster brake system offers superior operational performance and reliability for heavy-duty industrial crane applications. Therefore, replacing DC electromagnetic brakes with thruster brakes can significantly improve the overall performance and service life of EOT crane.
Keywords: EOT Crane, DC Electromagnetic Brake, Electro-Hydraulic Thruster Brake, Braking System, Performance Analysis, Material Handling Equipment, Crane Safety, Brake Modification, Industrial Automation, Brake Torque, Fail- Safe Braking, Mechanical Design.
The project involves studying the working principles of both braking systems and comparing their performance based on braking torque, response time, reliability, energy consumption, maintenance requirements, and operational safety. The design analysis includes load calculations, braking force estimation, motor compatibility, and stress considerations under different operating conditions. The thruster brake system provides fail-safe operation, faster response, better heat dissipation, and improved braking performance compared to the conventional DC electromagnetic brake.
The modification aims to enhance crane efficiency, reduce downtime, improve safety standards, and minimize maintenance costs. The results obtained from the analysis indicate that the electro-hydraulic thruster brake system offers superior operational performance and reliability for heavy-duty industrial crane applications. Therefore, replacing DC electromagnetic brakes with thruster brakes can significantly improve the overall performance and service life of EOT crane.
Keywords: EOT Crane, DC Electromagnetic Brake, Electro-Hydraulic Thruster Brake, Braking System, Performance Analysis, Material Handling Equipment, Crane Safety, Brake Modification, Industrial Automation, Brake Torque, Fail- Safe Braking, Mechanical Design.
Abstract
AI-Powered Multi-Purpose Agricultural UAV for Precision Farming, Smart Crop Protection, and Real-Time Monitoring: A Comprehensive Review
Nitin Narendra Ghanmode, Rutuja B M, Rudra Prakash Tiwari, Indra Prakash Tiwari
DOI: 10.17148/IJARCCE.2026.155169
Abstract: Agriculture is undergoing a major transformation with the adoption of intelligent technologies such as Artificial Intelligence (AI), Machine Learning (ML), Deep Learning (DL), remote sensing, and Unmanned Aerial Vehicles (UAVs). Traditional farming methods often require excessive labor, large quantities of pesticides, and continuous field monitoring, making agricultural operations time-consuming and inefficient. Precision agriculture has emerged as a modern solution to improve productivity, optimize resource utilization, and support sustainable farming practices [1], [3]. Among the various smart farming technologies, UAVs or agricultural drones have gained significant importance because of their capability to perform autonomous monitoring, crop analysis, precision spraying, and land surveying operations [12], [13].
This review paper presents a comprehensive study of an AI-powered multi-purpose agricultural UAV capable of performing fertilizer and pesticide spraying, crop health monitoring using AI-based image processing, fire detection and extinguishing, land surveying and GIS mapping, and animal/human detection for crop protection. The paper reviews recent advancements in UAV remote sensing technologies, multispectral and hyperspectral imaging systems, deep learning architectures, precision spraying mechanisms, and autonomous navigation systems used in precision agriculture [2], [21], [26].
Deep learning models such as Convolutional Neural Networks (CNNs), YOLO-based object detectors, Faster R-CNN, and Vision Transformers have significantly improved crop disease detection, object recognition, and aerial image analysis [18], [23], [27]. Remote sensing technologies including RGB imaging, thermal imaging, multispectral sensing, hyperspectral sensing, and LiDAR enable accurate vegetation monitoring, disease diagnosis, and terrain mapping [4], [24], [25]. Precision spraying systems integrated with AI-assisted decision-making improve spraying efficiency and reduce pesticide wastage [14], [29].
The review concludes that AI-powered multi-purpose agricultural UAVs provide an intelligent, efficient, and sustainable solution for next-generation smart farming applications. Continuous advancements in AI, sensor technologies, and autonomous UAV systems are expected to further revolutionize modern agriculture [13], [35].
Keywords: Precision Agriculture, UAV, Agricultural Drone, Artificial Intelligence, Machine Learning, Deep Learning, Remote Sensing, Smart Farming, Crop Disease Detection, YOLO, CNN, GIS, Precision Spraying.
This review paper presents a comprehensive study of an AI-powered multi-purpose agricultural UAV capable of performing fertilizer and pesticide spraying, crop health monitoring using AI-based image processing, fire detection and extinguishing, land surveying and GIS mapping, and animal/human detection for crop protection. The paper reviews recent advancements in UAV remote sensing technologies, multispectral and hyperspectral imaging systems, deep learning architectures, precision spraying mechanisms, and autonomous navigation systems used in precision agriculture [2], [21], [26].
Deep learning models such as Convolutional Neural Networks (CNNs), YOLO-based object detectors, Faster R-CNN, and Vision Transformers have significantly improved crop disease detection, object recognition, and aerial image analysis [18], [23], [27]. Remote sensing technologies including RGB imaging, thermal imaging, multispectral sensing, hyperspectral sensing, and LiDAR enable accurate vegetation monitoring, disease diagnosis, and terrain mapping [4], [24], [25]. Precision spraying systems integrated with AI-assisted decision-making improve spraying efficiency and reduce pesticide wastage [14], [29].
The review concludes that AI-powered multi-purpose agricultural UAVs provide an intelligent, efficient, and sustainable solution for next-generation smart farming applications. Continuous advancements in AI, sensor technologies, and autonomous UAV systems are expected to further revolutionize modern agriculture [13], [35].
Keywords: Precision Agriculture, UAV, Agricultural Drone, Artificial Intelligence, Machine Learning, Deep Learning, Remote Sensing, Smart Farming, Crop Disease Detection, YOLO, CNN, GIS, Precision Spraying.
Abstract
AUTONOMOUS AGRICULTURAL ROBOT FOR SEEDING, SPRAYING, AND WEEDING
Anosh Gundi, Anand Kulagod, Hebry Sunny K, Parashuram B
DOI: 10.17148/IJARCCE.2026.155170
Abstract: This project presents the design and development of an Autonomous Agricultural Robot for Seeding, Spraying, and Weeding operations aimed at reducing manual labour and improving farming efficiency. The system is built using a PIC Microcontroller, which acts as the main control unit to coordinate all operations of the robot. Utilizing a 1.5 mm square hollow mild steel frame with compact dimensions (24" x 16") to navigate small-scale fields, the movement is powered by PMDC worm gear motors and 8-inch nylon wheels, providing high torque for uneven terrains. The system integrates a 250g GI sheet seed box, a 500ml pesticide tank with a pump/nozzle, and a mild steel weeding mechanism. Operating in a semi-autonomous mode, this robot aims to reduce manual labor, minimize chemical wastage, and improve farming efficiency through a cost-effective, userfriendly design.
Keywords: Agricultural Robot, PIC Microcontroller, Seeding, Spraying, Mechanical Weeding, Precision Farming.
Keywords: Agricultural Robot, PIC Microcontroller, Seeding, Spraying, Mechanical Weeding, Precision Farming.
Abstract
A Hybrid EIGamal - AES Based Secure Image Encryption System
Gunipe Abhinav, K.V.V Subba Rao
DOI: 10.17148/IJARCCE.2026.155171
Abstract: The rapid progress of digital communication has increased the need for secure image protection mechanisms in applications such as healthcare, cloud storage, surveillance, military communication, and multimedia systems. Since image data contains high redundancy and strong correlation between neighbouring pixels, traditional text-oriented encryption techniques alone are not always sufficient for efficient and secure image protection. This paper presents a hybrid image encryption framework that combines the Advanced Encryption Standard (AES) and the ElGamal public- key cryptosystem to achieve both computational efficiency and secure key management. In the proposed model, AES is used to encrypt image data because of its fast processing capability, while ElGamal is employed to protect the AES secret key during transmission. The system also integrates statistical and performance evaluation modules to validate encryption strength. Security analysis is performed using entropy evaluation, histogram analysis, NPCR, UACI, key sensitivity testing, and noise attack simulation. Experimental observations show that the encrypted images exhibit strong randomness, uniform pixel distribution, and high resistance against differential and statistical attacks. The framework also demonstrates efficient encryption and decryption performance for different image sizes. The proposed hybrid approach provides a practical and reliable solution for secure image transmission and storage applications.
Keywords: Image Encryption, AES, ElGamal, Hybrid Cryptography, Secure Image Transmission, NPCR, UACI, Entropy Analysis
Keywords: Image Encryption, AES, ElGamal, Hybrid Cryptography, Secure Image Transmission, NPCR, UACI, Entropy Analysis
Abstract
Real-Time Hand Gesture Recognition and Sign Language Detection
Boda Deepthi, Tutta Naga Venkata Durga
DOI: 10.17148/IJARCCE.2026.155172
Abstract: This paper presents a real‑time system for recognizing American Sign Language gestures using deep learning. A webcam captures hand gestures, which are classified into 26 alphabets (A–Z) and three symbols (DEL, SPACE, NOTHING). The system employs a convolutional neural network with transfer learning from a pre‑trained ImageNet model, eliminating manual feature extraction. Implemented in Python using TensorFlow and OpenCV, the pipeline preprocesses live frames and performs instant gesture prediction. Experimental results demonstrate high classification accuracy and smooth real‑time performance under normal lighting conditions. This affordable, hardware‑free solution serves as an assistive tool to improve communication between hearing‑impaired individuals and non‑signers.
Keywords: American Sign Language recognition, Convolutional Neural Networks, transfer learning, real-time gesture classification, assistive communication technology, computer vision
Keywords: American Sign Language recognition, Convolutional Neural Networks, transfer learning, real-time gesture classification, assistive communication technology, computer vision
Abstract
Mechanical Steering-Controlled Adaptive Headlight System
Arbaz.U, Chandrashekhar.K, S. Immanviel David Raj, K. Durgashashank
DOI: 10.17148/IJARCCE.2026.155173
Abstract: Night-time driving accidents mainly occur due to poor road visibility during turns and curves. Conventional vehicle headlights remain fixed in one direction and fail to illuminate curved roads effectively. This limitation reduces driver visibility and increases accident risks during night driving. To overcome this problem, a Mechanical Steering-Controlled Adaptive Headlight System is proposed.
The proposed system mechanically connects the steering mechanism with the headlight assembly so that the headlights automatically rotate according to steering wheel movement. During turning operations, the headlights illuminate the curved path, thereby improving visibility and driver safety. The system consists of a steering linkage mechanism, rotating headlight assembly, return spring mechanism, and pivot arrangement.
The developed system improves road illumination, reduces blind spots, enhances driver confidence, and minimizes accident possibilities during night driving. The proposed design is economical, simple, reliable, and suitable for automobiles operating in urban and highway conditions.
Keywords: Adaptive Headlight System, Steering Control, Automobile Safety, Night Driving, Mechanical Linkage, Vehicle Lighting System.
The proposed system mechanically connects the steering mechanism with the headlight assembly so that the headlights automatically rotate according to steering wheel movement. During turning operations, the headlights illuminate the curved path, thereby improving visibility and driver safety. The system consists of a steering linkage mechanism, rotating headlight assembly, return spring mechanism, and pivot arrangement.
The developed system improves road illumination, reduces blind spots, enhances driver confidence, and minimizes accident possibilities during night driving. The proposed design is economical, simple, reliable, and suitable for automobiles operating in urban and highway conditions.
Keywords: Adaptive Headlight System, Steering Control, Automobile Safety, Night Driving, Mechanical Linkage, Vehicle Lighting System.
Abstract
REDUCTION OF WORK ROLL PEEL OFF INSTANCES IN HOT STRIP MILL
Nawal Kumar Dubey, H. Vinod Kumar, Vinay Kumar.H, Pradeepa.A
DOI: 10.17148/IJARCCE.2026.155174
Abstract: Hot Strip Mills (HSM) are widely used in steel manufacturing industries for producing steel strips with high dimensional accuracy and surface quality. During rolling operations, work rolls are subjected to severe thermal, mechanical, and frictional stresses. These stresses often lead to surface defects such as work roll peel-off, which causes strip surface damage, production loss, increased maintenance cost, and reduced roll life.
The present work focuses on the reduction of work roll peel-off instances in hot strip mills through process parameter optimization and preventive maintenance techniques. The study investigates the major causes of peel-off failures such as thermal fatigue, improper cooling, excessive rolling load, lubrication issues, and material defects. Various corrective measures including optimized cooling water flow, controlled rolling parameters, improved lubrication, and roll surface inspection methods were implemented.
Experimental observations indicate a considerable reduction in peel-off occurrences after implementing the proposed methods. The study demonstrates that proper roll management and process control can significantly improve roll life, strip quality, and overall mill productivity.
Keywords: Hot Strip Mill, Work Roll Peel-Off, Thermal Fatigue, Roll Failure, Steel Rolling, Surface Defects, Process Optimization.
The present work focuses on the reduction of work roll peel-off instances in hot strip mills through process parameter optimization and preventive maintenance techniques. The study investigates the major causes of peel-off failures such as thermal fatigue, improper cooling, excessive rolling load, lubrication issues, and material defects. Various corrective measures including optimized cooling water flow, controlled rolling parameters, improved lubrication, and roll surface inspection methods were implemented.
Experimental observations indicate a considerable reduction in peel-off occurrences after implementing the proposed methods. The study demonstrates that proper roll management and process control can significantly improve roll life, strip quality, and overall mill productivity.
Keywords: Hot Strip Mill, Work Roll Peel-Off, Thermal Fatigue, Roll Failure, Steel Rolling, Surface Defects, Process Optimization.
Abstract
“Finite Element Analysis of Connecting Rod Stresses, Deformation, and Material Optimization”
A Kiran Kumar, A Varun Kumar, B Lingaraju, Chetan G Gotur
DOI: 10.17148/IJARCCE.2026.155175
Abstract: The connecting rod is a vital component that converts reciprocating motion into rotary motion, enduring extreme compressive and tensile loads (Khudhair et al., 2024; Uddin et al., 2026). This study employs Finite Element Analysis to evaluate the structural integrity and performance of connecting rods under various material configurations, including forged steel, Aluminum 360, and advanced composites like Al-MWCNT (Chumbre, 2018; Uddin et al., 2026). Utilizing simulation software such as ANSYS and SolidWorks, the analysis focuses on minimizing mass while maintaining a high factor of safety to improve engine efficiency (Mishra et al., 2023; Muhammad et al., 2020). Results indicate that while forged steel provides high fatigue strength, Aluminum 360 exhibits significantly lower total deformation and a substantial mass reduction (Uddin et al., 2026). Furthermore, optimization techniques demonstrate that material substitution can lead to a weight reduction of up to 43.48% without exceeding material yield strength (Mishra et al., 2023). This research concludes that FEA is an essential tool for identifying critical stress concentrations and optimizing the geometric design of connecting rods for modern high-performance engines (Mihalache et al., 2014; Sun, 2006).
Keywords: Connecting rod, Finite Element Analysis, Stress analysis, Material optimization, Aluminum 360, Weight reduction, ANSYS, Von-Mises stress.
Keywords: Connecting rod, Finite Element Analysis, Stress analysis, Material optimization, Aluminum 360, Weight reduction, ANSYS, Von-Mises stress.
Abstract
AI SMART CAR
Nandu G, Dr Kavitha A S, Kushal C, Hithesh Gowda, Nishanth G S
DOI: 10.17148/IJARCCE.2026.155176
Abstract: Autonomous driving is an emerging technology that enables vehicles to operate with minimal human intervention using artificial intelligence. This project focuses on AI-based smart cars developed on small-scale platforms for testing and research. These systems use sensors such as cameras, IMUs, and distance sensors to perceive their surroundings. Machine learning techniques like deep learning and reinforcement learning help in decision-making and control. The architecture includes perception, planning, and control modules for efficient operation. One major challenge is transferring models from simulation to real-world environments. Despite limitations, small-scale smart cars provide a cost-effective platform for innovation. They play a significant role in advancing intelligent transportation systems.
Abstract
AI-Based Cyber Fraud Detection System
Mr. Harikrishna, M Shreya Reddy, Manasa V, Kirti M Maracharaddi
DOI: 10.17148/IJARCCE.2026.155177
Abstract: The need for this project is to develop an AI-based system that is capable of identifying cyber fraud, credit card fraud, UPI fraud, and online payment fraud, among others. The proposed system would utilize machine learning techniques that analyze user behavior and transaction patterns with a view to identifying any anomaly and something that is suspect. Indeed, this system is expected to keep track of all fraud activities that occur instantly, hence all activity that contradicts the usual activity of a particular user might be identified by this system instantly. The system proposed involves monitoring all digital transaction aspects related to credit card payments, UPI, and all online transaction activity, among others. Moreover, this system has the capability of learning from data, hence its accuracy improves with time due to its behavioral and pattern recognition feature. The end result of this project is expected to be a quick, reliable, and accurate system that offers instant notifications that cut down losses and restrict all unauthorized transactions instantly, hence making digital money much more secure, and users can trust financial institutions more.
Keywords: Cyber Fraud, Real-Time Transaction Monitoring, Machine Learning, Anomaly Detection, Digital Payments.
Keywords: Cyber Fraud, Real-Time Transaction Monitoring, Machine Learning, Anomaly Detection, Digital Payments.
Abstract
AUTOMATED EVALUATION OF PROGRAMMING ASSIGNMENTS
Moksud Alam Mallik, Nousheen Begum, Hafsa Yasmeen, Dr. Mujeeb Hasan
DOI: 10.17148/IJARCCE.2026.155179
Abstract: Grading programming assignments by hand can be laborious and prone to mistakes. Feedback with failing test cases is generated using current tools. Nevertheless, this approach is ineffective and yields insufficient outcomes. In this paper, we provide AUTOGRADER, a tool that, given a single reference implementation of the issue, automatically assesses whether programming assignments are right and generates counterexamples. Our tool looks for semantically distinct execution pathways between a student's submission and the reference implementation rather than counting the passed tests. The proposal is considered wrong if such a discrepancy is discovered; if not, it is considered to be the right answer. To capture the semantics of execution paths and identify possible path variations, we employ symbolic execution and weakest preconditions. AUTOGRADER is the first automated grading tool that relies on program semantics and generates feedback with counterexamples based on path deviations. It also reduces human efforts in writing test cases and makes the grading more complete. We implement AUTOGRADER and test its effectiveness and performance with real-world programming problems and student submissions collected from an online programming site. Our experiment reveals that there are no false negatives using our proposed method and we detected 11 errors of online platform judges.
Keywords: Automated Assessment, Programming Assignments, Software Testing, Automated Grading, Evaluation of code
Keywords: Automated Assessment, Programming Assignments, Software Testing, Automated Grading, Evaluation of code
Abstract
THERMOELECTRIC WASTE HEAT RECOVERY SYSTEM
Dharmana Prasad, P Sampath Kumar Achari, Penta Kunta Prudhvi, Praveen
DOI: 10.17148/IJARCCE.2026.155180
Abstract: Thermoelectric Waste Heat Recovery Systems (TWHRS) are gaining significant attention due to their ability to convert waste heat directly into electrical energy using thermoelectric generators (TEGs). In many industrial processes, automobiles, and power generation systems, a large amount of heat energy is lost to the environment. Recovering this wasted thermal energy can improve overall system efficiency and reduce fuel consumption and environmental pollution.
The present work focuses on the design and analysis of a thermoelectric waste heat recovery system using thermoelectric modules for converting exhaust heat into usable electrical power. The study investigates the major components, working principles, heat transfer mechanisms, and performance characteristics of the system. Parameters such as temperature difference, heat sink performance, module arrangement, and electrical output were analyzed.
Experimental observations indicate that increasing the temperature gradient across thermoelectric modules significantly improves voltage generation and power output. The developed system demonstrates improved energy efficiency and sustainable power generation from otherwise wasted thermal energy.
Keywords: Thermoelectric Generator, Waste Heat Recovery, Seebeck Effect, Energy Efficiency, Heat Recovery System, Sustainable Energy.
The present work focuses on the design and analysis of a thermoelectric waste heat recovery system using thermoelectric modules for converting exhaust heat into usable electrical power. The study investigates the major components, working principles, heat transfer mechanisms, and performance characteristics of the system. Parameters such as temperature difference, heat sink performance, module arrangement, and electrical output were analyzed.
Experimental observations indicate that increasing the temperature gradient across thermoelectric modules significantly improves voltage generation and power output. The developed system demonstrates improved energy efficiency and sustainable power generation from otherwise wasted thermal energy.
Keywords: Thermoelectric Generator, Waste Heat Recovery, Seebeck Effect, Energy Efficiency, Heat Recovery System, Sustainable Energy.
Abstract
Hybrid-Solar-Dryer
Muskan Begum, B Shambhulinga, P Pavan Kumar, V Sudheer
DOI: 10.17148/IJARCCE.2026.155181
Abstract: Post-harvest agricultural losses in developing regions often reach approximately 30% due to inadequate preservation facilities and the limitations of traditional open-air sun drying. This research presents the design and performance analysis of a hybrid solar dryer developed to provide continuous, high-efficiency drying regardless of weather conditions. The system integrates a flat-plate solar collector with an auxiliary biomass-fuelled gasifier heating unit, utilizing a 50-W photovoltaic panel to power fans for forced convection. Experimental results demonstrate that the hybrid system can achieve stable drying temperatures of up to 80°C with a thermal efficiency of approximately 35%. In performance trials, the dryer successfully reduced the moisture content of highly perishable produce—such as peaches and apples—from initial levels of ~89% to a safe storage level of 15–16% within 2 to 3 days. Furthermore, the integration of solar and biomass energy resulted in a 15% reduction in fuel consumption compared to biomass-only heating. This hybrid technology offers a sustainable, cost-effective, and hygienic solution for agricultural preservation, significantly improving product quality and shelf life in rural and industrial applications.
Keywords: Hybrid Solar Dryer, Biomass Heating, Post-Harvest Loss, Thermal Efficiency, Renewable Energy, Agricultural Preservation, Forced Convection, Moisture Content, Solar Thermal, Sustainable Agriculture, Food Security, Biomass Gasifier, Dehydration Technology, Rural Development.
Keywords: Hybrid Solar Dryer, Biomass Heating, Post-Harvest Loss, Thermal Efficiency, Renewable Energy, Agricultural Preservation, Forced Convection, Moisture Content, Solar Thermal, Sustainable Agriculture, Food Security, Biomass Gasifier, Dehydration Technology, Rural Development.
Abstract
A Smart AI-Based Food Redistribution System for Reducing Food Waste and Supporting Community Welfare
Dr. Phanindra Reddy K, Kalyani N T, Kavya B, Keerthana P, P Sahana
DOI: 10.17148/IJARCCE.2026.155182
Abstract: The increasing rate of food wastage and food insecurity has become a major global concern affecting environmental sustainability and social welfare. Large amounts of edible food produced by restaurants, hotels, functions, and households are discarded daily due to the absence of an efficient redistribution mechanism. To address this issue, the AI-Powered Residual Food Distribution System is proposed as an intelligent platform that automates the collection, analysis, and redistribution of surplus food to needy individuals and organizations. The proposed system integrates Artificial Intelligence, machine learning, image processing, geolocation services, and web technologies to create a centralized and efficient food redistribution platform. Donors can upload surplus food details along with food images, while the AI mod- ule analyzes food quality and determines whether the food is safe and suitable for distribution. Based on location and availability, the system automatically identifies nearby NGOs and volun- teers and sends real-time notifications for food pickup and delivery. The system also incorporates secure authentication, centralized database management, GPS-based route optimization, and real-time tracking to improve operational efficiency and transparency. Experimental analysis shows that the proposed system significantly reduces food spoilage, improves coordination among stakeholders, minimizes wastage, and ensures timely food delivery. Overall, the system provides a smart and sustainable technological solution for efficient food redistribution and community welfare.
Keywords: Artificial Intelligence, Food Waste Management, Smart Food Distribution, Machine Learning, Food Donation System, Food Quality Detection, Geolocation Services, Web Application
Keywords: Artificial Intelligence, Food Waste Management, Smart Food Distribution, Machine Learning, Food Donation System, Food Quality Detection, Geolocation Services, Web Application
Abstract
Multidisciplinary Review on FEA, Topology Optimization and AI-Based Analysis of Aero-Engine Turbine Blades
Zayed Mulla, Mujaffar Hussain, Y. V. Rohinish, Mohammed Aayyon Khan
DOI: 10.17148/IJARCCE.2026.155183
Abstract: The design and analysis of turbine blades for aero-engines, especially those in the high-pressure turbine stage of turbofan engines, represent some of the most intricate challenges in aerospace engineering due to the intense thermal, aerodynamic, and mechanical stresses they face. Recent developments in computational methods, additive manufacturing, topology optimization, and artificial intelligence have significantly propelled research in turbine blades forward. This review paper delves into notable advancements in areas such as turbine blade modeling, computational fluid dynamics (CFD), finite element analysis (FEA), fatigue analysis, material optimization, lattice structures [2], [3], [4], fluid-structure interaction [13], and AI-driven defect detection [1]. The review is based on an analysis of twenty- three research papers from journals, conferences, and reports that concentrate on gas turbine blade performance design and CAE (Computer Aided Engineering) applications. The gathered studies were organized according to aerodynamic analysis [5], [6], [7], [8], thermal analysis, structural analysis, topology optimization, vibration behavior, and material selection. The literature review reveals that CFD and FEA continue to be the predominant tools for forecasting stress, deformation, heat transfer, and aerodynamic performance in turbine blades. Recent research has highlighted that topology optimization [1], [9] and lattice-based internal structures can significantly decrease blade weight while enhancing structural integrity and vibration resistance. Moreover, deep learning techniques [1], [10] have shown great promise for the automated detection of defects and predictive maintenance of aero-engine blades. The review's findings indicate that future turbine blade systems are likely to incorporate lightweight structures, additive manufacturing, advanced cooling techniques, and AI-assisted monitoring systems. This paper offers a comprehensive understanding of current research trends and pinpoints future opportunities for interdisciplinary research on turbine blades.
Keywords: Gas Turbine Blades, Finite Element Analysis, Topology Optimisation, Fluid Structure Interaction.
Keywords: Gas Turbine Blades, Finite Element Analysis, Topology Optimisation, Fluid Structure Interaction.
Abstract
Medical Diagnosis for Coronary Arteries Disease Detection Using Deep Learning
J Aakash, M Balaji, M Syed Hussain, Dr. A Samuel Chelladhurai
DOI: 10.17148/IJARCCE.2026.155184
Abstract: Coronary Artery Disease (CAD) is among the most widespread and life-threatening cardiovascular conditions worldwide, responsible for a significant proportion of global mortality. Early and accurate detection of CAD plays a crucial role in enabling timely medical intervention and reducing the risk of severe cardiac complications. Conventional diagnostic approaches rely heavily on manual clinical analysis performed by experienced cardiologists, which is time- consuming, expensive, and susceptible to human error. This paper presents a Deep Learning-Based Coronary Artery Disease Detection System using a hybrid Convolutional Neural Network and Artificial Neural Network (CNN-ANN) architecture integrated with a Flask-based clinical web interface. The proposed system analyzes patient medical parameters from the UCI Heart Disease Dataset including age, blood pressure, cholesterol level, ECG reports, maximum heart rate, chest pain type, fasting blood sugar, and exercise-induced angina. The dataset undergoes systematic preprocessing involving normalization, label encoding, and an 80:20 train-test split. The Deep Learning model is implemented using TensorFlow and Keras with Dense layers, ReLU and Sigmoid activation functions, Dropout regularization, and Adam optimizer. A complementary module, CAD Vision Pro, extends the system to image-based coronary analysis using CCTA imaging with Grad-CAM heatmap visualization, risk scoring, severity classification, and automated PDF report generation. Experimental results demonstrate that the proposed system achieves approximately 95% prediction accuracy with high precision, recall, and F1-score values. The system provides healthcare professionals with an intelligent, automated, and user-friendly CAD detection platform that supports early diagnosis, reduces manual workload, and contributes toward AI-driven clinical decision support in modern healthcare.
Keywords: Coronary Artery Disease, Deep Learning, CNN, ANN, TensorFlow, Keras, Flask, UCI Heart Disease Dataset, CAD Vision Pro, Medical Image Analysis, Healthcare AI, CCTA, Grad-CAM
Keywords: Coronary Artery Disease, Deep Learning, CNN, ANN, TensorFlow, Keras, Flask, UCI Heart Disease Dataset, CAD Vision Pro, Medical Image Analysis, Healthcare AI, CCTA, Grad-CAM
Abstract
Human Following Robotics
K Judah Benhur, D C Darshan Gouda, D Bhanuchandra, H Chandu.
DOI: 10.17148/IJARCCE.2026.155185
Abstract: The development of human-following robots is a critical area within human-robot interaction, with applications in assistive healthcare, industrial logistics, and social companionship (Eirale et al., 2025; Islam et al., 2019). This paper presents the design of an autonomous mobile robot capable of real-time target tracking and following. The proposed system utilizes a multi-sensor approach—integrating RGB-D cameras for person detection and ultrasonic sensors for precise distance estimation (Fung et al., 2025; Rashid et al., 2012). The methodology leverages the Robot Operating System framework to coordinate perception and control (Pavlova & Bahrami, 2025; Priyandoko et al., 2018). Experimental results indicate that the robot can maintain a consistent following distance of approximately 1.3 meters and adapt to walking speeds of up to 0.7 m/s (Amaya et al., 2024; Gockley et al., 2007). This research addresses key challenges in perception and navigation, providing a foundation for reliable service robots in human-centric environments (Eirale et al., 2025).
Keywords: Human-following robot, Human-robot interaction, Autonomous navigation, Robot Operating System, Person tracking, Sensor fusion, Fuzzy logic control, Obstacle avoidance.
Keywords: Human-following robot, Human-robot interaction, Autonomous navigation, Robot Operating System, Person tracking, Sensor fusion, Fuzzy logic control, Obstacle avoidance.
Abstract
SOCIAL ENGINEERING ATTACK
Mohammad Salman, Thanuja J C
DOI: 10.17148/IJARCCE.2026.155186
Abstract: Social Engineering Attacks have become one of the most dangerous threats in the field of cybersecurity. Unlike traditional cyberattacks that target software vulnerabilities, social engineering focuses on manipulating human behavior to gain unauthorized access to sensitive information, systems, or networks. Attackers use psychological techniques such as fear, trust, urgency, and curiosity to deceive users into revealing confidential data like passwords, bank details, or personal information. Common forms of social engineering attacks include phishing, baiting, pretexting, tailgating, and vishing.
The increasing use of digital platforms, online banking, social media, and cloud services has significantly increased the risk of social engineering attacks. Traditional security systems such as firewalls and antivirus software cannot fully prevent these attacks because they target human weaknesses instead of technical flaws. Therefore, awareness and user education play an important role in reducing cyber risks.
This project presents a detailed study of social engineering attacks, their techniques, impacts, prevention methods, and cybersecurity awareness strategies. The system also analyzes how attackers manipulate victims and how organizations can strengthen security through training, authentication mechanisms, and monitoring systems. The study aims to create awareness about cyber threats and promote safe digital practices among users and organizations.
Keywords: Cybersecurity, Social Engineering, Phishing, Vishing, Baiting, Pretexting, Cyber Attacks, Information Security, Human Vulnerability, Authentication, Malware, Cyber Awareness, Digital Security, Identity Theft, Online Fraud, Network Security.
The increasing use of digital platforms, online banking, social media, and cloud services has significantly increased the risk of social engineering attacks. Traditional security systems such as firewalls and antivirus software cannot fully prevent these attacks because they target human weaknesses instead of technical flaws. Therefore, awareness and user education play an important role in reducing cyber risks.
This project presents a detailed study of social engineering attacks, their techniques, impacts, prevention methods, and cybersecurity awareness strategies. The system also analyzes how attackers manipulate victims and how organizations can strengthen security through training, authentication mechanisms, and monitoring systems. The study aims to create awareness about cyber threats and promote safe digital practices among users and organizations.
Keywords: Cybersecurity, Social Engineering, Phishing, Vishing, Baiting, Pretexting, Cyber Attacks, Information Security, Human Vulnerability, Authentication, Malware, Cyber Awareness, Digital Security, Identity Theft, Online Fraud, Network Security.
Abstract
REMOTE CONTROLLER PORTABLE DRILLING MACHINE
Pruthvi Raj RM, Rohit Sardar, Sundeep BM, Somushakar Reddy
DOI: 10.17148/IJARCCE.2026.155187
Abstract; Portable drilling machines are widely used in construction, metal fabrication, mining, and maintenance industries for drilling operations. However, conventional drilling machines require direct human operation, which may expose workers to dangerous environments and reduce operational flexibility. To overcome these limitations, this paper proposes a Remote Controller Portable Drilling Machine that enables drilling operations through wireless remote control technology.
The proposed system integrates a portable drilling mechanism with a wireless control unit, DC motor drive system, rechargeable battery, and microcontroller-based automation. The drilling machine can be operated remotely using RF/Bluetooth communication, allowing the operator to control movement, drilling speed, and drilling direction from a safe distance.
The system improves operator safety, portability, operational efficiency, and drilling accuracy. The lightweight structure and compact design make the machine suitable for industrial maintenance, construction sites, railway applications, hazardous environments, and remote drilling operations.
Keywords: Portable Drilling Machine, Remote Control, Wireless Operation, DC Motor, Automation, Microcontroller, Industrial Safety.
Abstract
PRODUCTION OF FUEL OUT OF PLASTIC WASTE USING PYROLYSIS PROCESS
Prajwal L, Jayashankar, G Sandeep, GV Varun
DOI: 10.17148/IJARCCE.2026.155188
Abstract: Plastic waste has become one of the major environmental problems across the world.
This project focuses on converting plastic waste into useful fuel through the pyrolysis process.
Pyrolysis is a thermal decomposition process carried out in the absence of oxygen, where plastic materials are heated at high temperatures to produce liquid fuel, gas, and char.
The produced fuel can be used as an alternative energy source in industries and engines after proper refining. This method reduces plastic pollution, minimizes landfill waste, and supports sustainable energy production.
Keywords: Plastic Waste, Pyrolysis, Fuel Production, Recycling, Alternative Energy
This project focuses on converting plastic waste into useful fuel through the pyrolysis process.
Pyrolysis is a thermal decomposition process carried out in the absence of oxygen, where plastic materials are heated at high temperatures to produce liquid fuel, gas, and char.
The produced fuel can be used as an alternative energy source in industries and engines after proper refining. This method reduces plastic pollution, minimizes landfill waste, and supports sustainable energy production.
Keywords: Plastic Waste, Pyrolysis, Fuel Production, Recycling, Alternative Energy
Abstract
Design & Fabrication of Bio-Composites Material for Automotive / Engineering Application
Abhishek Rajbhar, Suresh B, H P J Aalibaaba, Chandra Shekar Reddy K
DOI: 10.17148/IJARCCE.2026.155189
Abstract: The automotive and engineering sectors are undergoing a significant transition toward sustainability, driven by the need to reduce carbon footprints and improve fuel efficiency through vehicle lightweighting [1], [2]. This paper investigates the design and fabrication of biocomposite materials, which integrate natural fiber reinforcements—such as hemp, bamboo, jute, and fique—into various polymer matrices [3], [4]. We evaluate several manufacturing techniques, specifically compression molding, injection molding, and hand lay-up, which are selected based on production volume and geometric complexity [5], [6], [7].
Our findings indicate that these materials can achieve mechanical properties comparable to traditional synthetic composites; for instance, fique-epoxy biocomposites have demonstrated tensile strengths of approximately 36.6 MPa and flexural strengths of 21.2 MPa [3]. Furthermore, bio-hybrid structures using hemp and bamboo have exhibited impact energies as high as 18.33 J, significantly outperforming standard materials [4]. Despite challenges related to moisture absorption and fiber-matrix adhesion, the high thermal stability (up to 430.6 °C) and carbon-neutral lifecycle of these composites make them essential for the future of "green" manufacturing and industrial recyclability [4], [7], [8].
Keywords: Biocomposites; Natural Fibers; Automotive Engineering; Sustainable Materials; Lightweighting; Mechanical Characterization; Green Manufacturing; Polymer Matrices.
Our findings indicate that these materials can achieve mechanical properties comparable to traditional synthetic composites; for instance, fique-epoxy biocomposites have demonstrated tensile strengths of approximately 36.6 MPa and flexural strengths of 21.2 MPa [3]. Furthermore, bio-hybrid structures using hemp and bamboo have exhibited impact energies as high as 18.33 J, significantly outperforming standard materials [4]. Despite challenges related to moisture absorption and fiber-matrix adhesion, the high thermal stability (up to 430.6 °C) and carbon-neutral lifecycle of these composites make them essential for the future of "green" manufacturing and industrial recyclability [4], [7], [8].
Keywords: Biocomposites; Natural Fibers; Automotive Engineering; Sustainable Materials; Lightweighting; Mechanical Characterization; Green Manufacturing; Polymer Matrices.
Abstract
MONKEYPOX SKIN LESION
Bhoomika Barki, Vidya S
DOI: 10.17148/IJARCCE.2026.155190
Abstract: Early detection of infectious skin diseases such as Monkeypox, Chickenpox, and Measles is essential for preventing complications and reducing the spread of infection. However, access to dermatologists and advanced diagnostic facilities is limited in many regions, making timely diagnosis difficult. This project presents an intelligent web-based Skin Disease Detection System that uses deep learning techniques to automatically classify skin images into Monkeypox, Chickenpox, Measles, or Normal categories. The system employs a Convolutional Neural Network (CNN) model trained on skin image datasets to perform accurate image-based classification. Users can upload skin images through a web interface, and the system processes the input to generate prediction results along with confidence scores and basic medical information. A secure database is used to store user details and prediction history, while automated email notifications are sent to users for report confirmation. An admin or doctor dashboard is included for monitoring system activity and analysing disease trends. Experimental evaluation shows that the system provides reliable predictions, improves accessibility to preliminary diagnosis, and supports early disease awareness, making it a useful decision-support tool for healthcare applications.
Keywords: Skin Disease Detection, Deep Learning, Convolutional Neural Network, Medical Image Analysis, Web- Based Healthcare System, Automated Diagnosis
Keywords: Skin Disease Detection, Deep Learning, Convolutional Neural Network, Medical Image Analysis, Web- Based Healthcare System, Automated Diagnosis
Abstract
OPTIMIZATION OF NITROGEN TO DRY COAL BUNKERS
P. Vijay Kumar, P. Sandeepa, Tarun Kumar BC, K. Sudarshana
DOI: 10.17148/IJARCCE.2026.155191
Abstract: Coal bunkers are widely used in thermal power plants for storing pulverized coal before combustion. Moisture present inside coal bunkers affects combustion efficiency, causes coal sticking, corrosion, and increases the risk of spontaneous combustion. Conventional drying methods consume high energy and are often inefficient for maintaining safe moisture levels.
This paper proposes the optimization of nitrogen gas for drying coal bunkers to improve coal handling efficiency, reduce moisture content, and enhance operational safety. Nitrogen is used as an inert drying medium because of its non-reactive nature and ability to minimize oxygen concentration inside bunkers. The proposed system utilizes controlled nitrogen flow, humidity monitoring sensors, pressure regulators, and automated control systems to achieve efficient drying with minimum nitrogen consumption.
The study focuses on improving drying efficiency, reducing coal degradation, preventing fire hazards, and minimizing operational costs. Experimental analysis shows that optimized nitrogen circulation significantly reduces bunker moisture while maintaining safe environmental conditions. The proposed method is suitable for modern thermal power plants requiring safe and efficient coal storage systems.
Keywords: Coal Bunker, Nitrogen Drying, Thermal Power Plant, Moisture Reduction, Inert Gas System, Coal Storage Safety, Optimization.
This paper proposes the optimization of nitrogen gas for drying coal bunkers to improve coal handling efficiency, reduce moisture content, and enhance operational safety. Nitrogen is used as an inert drying medium because of its non-reactive nature and ability to minimize oxygen concentration inside bunkers. The proposed system utilizes controlled nitrogen flow, humidity monitoring sensors, pressure regulators, and automated control systems to achieve efficient drying with minimum nitrogen consumption.
The study focuses on improving drying efficiency, reducing coal degradation, preventing fire hazards, and minimizing operational costs. Experimental analysis shows that optimized nitrogen circulation significantly reduces bunker moisture while maintaining safe environmental conditions. The proposed method is suitable for modern thermal power plants requiring safe and efficient coal storage systems.
Keywords: Coal Bunker, Nitrogen Drying, Thermal Power Plant, Moisture Reduction, Inert Gas System, Coal Storage Safety, Optimization.
Abstract
NAMMA BUS: Intelligent Transportation System Using Deep Learning
Dr. B. M. Vidyavathi, Suchitra M, V Rakshitha, Yamini V, Eshwar G
DOI: 10.17148/IJARCCE.2026.155192
Abstract: efficient transportation management plays an important role in educational institutions, where students and faculty members depend on campus buses for daily commuting. Traditional transportation systems often face challenges such as uncertain bus arrival times, traffic delays, lack of real-time tracking, and ineffective communication between drivers and passengers. These issues can lead to longer waiting times, missed buses, and reduced transportation efficiency. To address these limitations, this paper presents NAMMA BUS, an intelligent transportation system designed using Internet of Things (IoT), Global Positioning System (GPS), and Deep Learning-based Regression techniques. The proposed system collects real-time data such as bus location, speed, route information, and traffic conditions using GPSenabled devices installed in buses. The collected data is transmitted to a centralized server through IoT modules for processing and analysis. A Deep Learning Regression model is implemented to predict the Estimated Time of Arrival (ETA) by analyzing historical travel records and live movement patterns. The system follows a three-tier architecture, enabling students to track buses through mobile or web applications, while drivers and administrators manage transportation activities through dedicated dashboards. Experimental results show improved prediction accuracy, reduced passenger waiting time, and enhanced operational transparency, making the system a reliable solution for smart campus transportation.
Keywords: Intelligent transportation systems, deep learning, traffic prediction, vehicle detection, smart traffic management.
Keywords: Intelligent transportation systems, deep learning, traffic prediction, vehicle detection, smart traffic management.
Abstract
A Comprehensive Review on IoT-Based Smart Agriculture Systems
Sameer Sheikh
DOI: 10.17148/IJARCCE.2026.155193
Abstract: The rapid growth in global population has significantly increased the demand for efficient and sustainable agricultural practices. Traditional farming approaches often suffer from limitations such as inefficient water usage, lack of real-time monitoring, and heavy reliance on manual labor. The Internet of Things (IoT) has emerged as a transformative technology capable of addressing these challenges by enabling smart and automated agricultural systems [1], [2]. This paper presents a comprehensive review of IoT-based smart agriculture systems. It examines various components including sensors, communication technologies, cloud platforms, and smart irrigation mechanisms. The study also evaluates multiple research contributions to highlight improvements in productivity, resource utilization, and decision- making processes. Furthermore, key challenges such as infrastructure cost, network reliability, and system maintenance are discussed. Finally, future research directions involving artificial intelligence, machine learning, and predictive analytics are explored to enhance precision farming.
Keywords: IoT, Smart Agriculture, Precision Farming, Sensor Networks, Smart Irrigation
Keywords: IoT, Smart Agriculture, Precision Farming, Sensor Networks, Smart Irrigation
Abstract
A Survey on Machine Learning Approach for Momentum Shift Detection and Win Prediction in Cricket Matches
Mr. J. R. Harshavardhan, Mithun M Parashar, Pavan D R, Punith Gowda G, and, Sachin N B
DOI: 10.17148/IJARCCE.2026.155194
Abstract: Cricket is among the most strategically complex and data-intensive sports in the world, producing extensive real-time performance data across multiple formats. Despite significant advances in sports analytics, existing research predominantly focuses on pre-match outcome prediction or score estimation, with limited investigation into in-match momentum dynamics and their effect on win probability. This survey critically examines five state-of-the-art contributions: momentum shift modeling in sports [1], DoE and regression-based cricket analytics [2], ensemble ML- based outcome classification [3], T20 dangerous-ball impact analysis [4], and ODI score prediction using regression [5]. Through systematic comparative analysis employing three reference tables, seven critical research gaps are identified— most notably the complete absence of real-time cricket-specific momentum shift detection frameworks. In response, this paper proposes an intelligent system integrating LSTM-based temporal momentum detection, XGBoost ensemble win probability prediction, a novel Cricket Momentum Index (CMI), and an interactive Real time analytical dashboard covering all cricket formats
Keywords: Cricket Analytics; Momentum Shift Detection; Win Probability Prediction; Machine Learning; LSTM; XGBoost; LightGBM; Random Forest; Gradient Boosting; Real-Time Sports Analytics; Match Outcome Prediction; IPL; ODI; T20;Feature engineering
Keywords: Cricket Analytics; Momentum Shift Detection; Win Probability Prediction; Machine Learning; LSTM; XGBoost; LightGBM; Random Forest; Gradient Boosting; Real-Time Sports Analytics; Match Outcome Prediction; IPL; ODI; T20;Feature engineering
Abstract
Secure Real-Time Coding Interview System with AI-Based Evaluation and Monitoring
Mr. J.R. Harshavardhan, Shreyas B, Yashas A, Yashwanth K, and Shiva Prakash J
DOI: 10.17148/IJARCCE.2026.155195
Abstract: The growing demand for remote technical hiring has accelerated the need for intelligent, unified platforms capable of conducting fair, scalable, and secure coding interviews. Traditional tools address only isolated aspects— collaboration, execution, or evaluation—leaving critical gaps in assessment integrity and real-time interactivity. This paper presents a Secure Real-Time Coding Interview System that integrates a conflict-free multi-user collaborative code editor, AI-based automated performance evaluation, Docker-sandboxed secure code execution, and behavioral monitoring into a single cloud-based platform. The system leverages WebSockets and CRDT/OT algorithms for real- time synchronization, Judge0 API for multi-language execution, and WebRTC/Socket.io for voice, video, and chat communication. Through a systematic review of five related works, six critical research gaps are identified, and a five- layer architecture is proposed to address them comprehensively. The platform requires no local installation and targets academic institutions and recruitment organizations seeking credible, interactive, and scalable online coding assessment infrastructure.
Keywords: Real-Time Collaboration; Coding Interview; AI Evaluation; WebSockets; CRDT; Docker; Monaco Editor; Secure Code Execution; Online Assessment; Socket.io; WebRTC; Node.js; React.js; Judge0 API; Behavioral Monitoring.
Keywords: Real-Time Collaboration; Coding Interview; AI Evaluation; WebSockets; CRDT; Docker; Monaco Editor; Secure Code Execution; Online Assessment; Socket.io; WebRTC; Node.js; React.js; Judge0 API; Behavioral Monitoring.
Abstract
Smart Civic AI: Intelligent Complaint Management System
Archana N, Yogeshwaran G, Uday D R, Manoj Kumar C, Sachin R
DOI: 10.17148/IJARCCE.2026.155196
Abstract: Urban civic complaint systems continue to face inefficiencies such as delayed responses, lack of transparency, manual routing errors, and poor citizen engagement. This paper presents a survey of existing complaint management systems and proposes Smart Civic AI: Intelligent Complaint Management System, an advanced smart governance platform that integrates Artificial Intelligence, image recognition, voice-based complaint registration (IVR), real-time GPS navigation, heatmap visualization, and priority-based complaint management with role-based dashboards. Citizens can report issues such as potholes, sewage overflow, garbage accumulation, water leakage, and streetlight failures by capturing live images or through voice input. AI automatically classifies the issue, generates complaint descriptions, detects duplicate complaints, and assigns priority levels based on severity and location. Complaints are routed to the relevant authority dashboard, while workers receive navigation support using OpenStreetMap API and update work progress in real time. The system also generates complaint heatmaps to identify high-density problem areas for better resource allocation and smart city planning. Real-time status updates are reflected across citizen and authority dashboards, improving transparency and accountability. The proposed system addresses major limitations identified in existing research works, including lack of AI-driven classification, absence of voice-enabled complaint registration, no worker navigation support, limited analytics, and lack of intelligent escalation mechanisms, thereby enhancing the efficiency and effectiveness of smart city governance.
Keywords: Complaint Management System, AI Image Recognition, IVR, Smart City, GPS Navigation, OpenStreetMap, Heatmap Analysis, Priority-Based Complaints, Role-Based Dashboard, Civic Engagement.
Keywords: Complaint Management System, AI Image Recognition, IVR, Smart City, GPS Navigation, OpenStreetMap, Heatmap Analysis, Priority-Based Complaints, Role-Based Dashboard, Civic Engagement.
Abstract
An Explainable AI Framework for Lifestyle-Based Healthcare Prediction: A Comprehensive Survey
Vidyasre N, Mahit Rao P, Mithun N, S G Ravidas, Shashidhara S C
DOI: 10.17148/IJARCCE.2026.155197
Abstract: Artificial Intelligence (AI) has become a driving force in preventive healthcare, enabling data-driven prediction and early detection of diseases. However, black-box models often lack interpretability, limiting clinical trust and practical adoption. This survey presents a comprehensive analysis of recent Explainable AI (XAI)-driven healthcare prediction systems focusing on lifestyle-based risk assessment. The study reviews six prominent works integrating machine learning (ML), deep learning, and XAI methods such as Shapley Additive Explanations (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME) for disease detection—including cardiometabolic conditions, diabetes, heart disease, and mental health risk assessment. Key findings emphasise that explanation consistency, data imbalance, and lack of multimodal integration remain open challenges. This paper consolidates methodologies, compares architectural frameworks, identifies research gaps, and proposes a unified model for lifestyle-based multi-disease prediction using interpretable ML. The proposed architecture integrates SHAP and LIME with Random Forest and Logistic Regression to generate transparent predictions, enhancing clinical usability and preventive intervention design.
Keywords: Explainable Artificial Intelligence, Healthcare Analytics, Lifestyle-Based Prediction, SHAP, LIME, Machine Learning, Preventive Healthcare, Disease Risk Prediction
Keywords: Explainable Artificial Intelligence, Healthcare Analytics, Lifestyle-Based Prediction, SHAP, LIME, Machine Learning, Preventive Healthcare, Disease Risk Prediction
Abstract
“HEAT TRANSFER ANALAYSIS AND OPTIMIZATIION OF FINS BY VARIATION IN GEOMETRY”
Siddalinga K, Pinjari Asif, Murali Krishana K, C D Mounesh
DOI: 10.17148/IJARCCE.2026.155198
Abstract: This paper explores the thermal performance and geometric optimization of extended surfaces (fins) in heat dissipation applications. By evaluating various geometries—including rectangular, triangular, parabolic, and pin fins— we analyze the trade-offs between heat transfer rates and material volume. Modern optimization techniques, such as Random Forest metamodels and multi-fidelity evolutionary algorithms, are discussed as means to reduce material consumption while maintaining thermal reliability. Key findings indicate that while rectangular fins provide high absolute heat flux, triangular and parabolic profiles offer superior material efficiency. Numerical studies show that intelligent optimization can reduce heat sink volume by up to 26% with minimal temperature penalties (Sun et al., 2025).
Abstract
Engineering Students’ Perception of AI Replacing Human Skills in Future Technical Careers: A Survey Study
Shreya.S, Tanushree. R, Srinidhi.N, Dr. Sonia Maria D'souza
DOI: 10.17148/IJARCCE.2026.155199
Abstract: The survey looked at what students know about what Artificial Intelligence can do. It also looked at how confident students are in things that only humans can do, such as thinking critically, being creative, making good judgments, and working well with others. The survey wanted to see what students think about Artificial Intelligence taking jobs versus Artificial Intelligence helping with jobs. Early results show that although most students agree with the improving abilities of AI in performing monotonous and analytical tasks, a great number are confident about the irreplaceable nature of complex human mental and emotional abilities. Nonetheless, there is some concern about job displacement in occupations requiring highly repetitive technical skills. There is also an observable relationship between AI knowledge and optimistic career prospects, indicating that AI instruction could.
Keywords: Artificial Intelligence, Engineering Students, Career Perception, Skill Displacement, AI Literacy, Future of Work, Human Skills, Career Anxiety, Technical Education, Workforce Disruption
Keywords: Artificial Intelligence, Engineering Students, Career Perception, Skill Displacement, AI Literacy, Future of Work, Human Skills, Career Anxiety, Technical Education, Workforce Disruption
Abstract
STREET CARE OPERATION PORTAL
Abhijith, Theerthashree G S
DOI: 10.17148/IJARCCE.2026.155200
Abstract: The rapid growth of urban areas has increased the demand for efficient systems to manage civic issues and public infrastructure complaints. Traditional complaint management approaches often depend on manual procedures, phone calls, or fragmented digital systems, resulting in delayed response times, lack of transparency, and poor coordination between authorities and field workers. In recent years, modern web technologies and intelligent digital platforms have emerged as effective solutions for improving civic issue reporting and operational management. This paper presents a study on the application of the Street Care Operation Portal for centralized civic issue reporting, complaint management, and worker coordination. The proposed system enables citizens to report issues such as damaged roads, drainage blockages, waste accumulation, and streetlight failures by providing descriptions, media evidence, and live location information. The system architecture follows a modular client–server design where the backend manages authentication, issue processing, worker assignment, and data storage, while the frontend provides interactive dashboards for citizens, administrators, and field workers. Role-based access control, location-based services, and real-time status tracking improve operational transparency and accountability. Administrators can monitor reported issues, assign tasks to workers, and analyze system activity through visual dashboards, while field workers update issue resolution progress directly through the platform. The results demonstrate that the Street Care Operation Portal improves coordination, reduces manual effort, and enhances response efficiency compared to traditional civic complaint systems. The application is implemented as a web-based system using React.js on the frontend with a Node.js and Express.js backend, and MongoDB for data storage.
Keywords: Smart City Management, Civic Issue Reporting, Complaint Management System, MERN Stack, React.js, Node.js, Express.js, MongoDB, JWT Authentication, Role-Based Access Control, Web Application, OpenStreetMap Integration, Real-Time Status Tracking, Admin Dashboard, Field Worker Management, Geolocation Services, RESTful APIs, Urban Infrastructure Management, Civic Complaint Portal, Analytics Dashboard, Full-Stack Development, Digital Governance, Smart Civic Services.
Keywords: Smart City Management, Civic Issue Reporting, Complaint Management System, MERN Stack, React.js, Node.js, Express.js, MongoDB, JWT Authentication, Role-Based Access Control, Web Application, OpenStreetMap Integration, Real-Time Status Tracking, Admin Dashboard, Field Worker Management, Geolocation Services, RESTful APIs, Urban Infrastructure Management, Civic Complaint Portal, Analytics Dashboard, Full-Stack Development, Digital Governance, Smart Civic Services.
Abstract
AI-Powered Smart Biker Safety and Monitoring Platform
Dakshayini G R, Megha, Veena S Malipatil, Vindhya R
DOI: 10.17148/IJARCCE.2026.155201
Abstract: Accidents causing severe injuries are mostly caused by two-wheelers, with inadequate response to emergencies being another issue here. Several factors cause this problem, including failure to wear helmets, accident monitoring, and reporting. Conventional approaches to solving this problem rely heavily on human intervention such as human-based monitoring and rider vigilance. It may be insufficient in case of emergencies.
In this context, AI-Based Smart Bike Safety and Accident Detection System Using IoT and Deep Learning will be proposed. This approach uses Artificial Intelligence to monitor helmet use, detect any accidents, track location and provide alert when necessary. Helmet detection is used to ensure that the rider is using the helmet before operating the motorcycle. Sensor detection detects the occurrence of accidents through the detection of collision, tilt, and stationary. In case of an accident, the system notifies people and gives the location of the accident.
This paper discusses the implementation of AI-based smart bike safety and accident detection system.
Keywords: Smart bike safety, helmet detection, accident detection, IoT, deep learning, GPS tracking, emergency alert system, rider safety.
In this context, AI-Based Smart Bike Safety and Accident Detection System Using IoT and Deep Learning will be proposed. This approach uses Artificial Intelligence to monitor helmet use, detect any accidents, track location and provide alert when necessary. Helmet detection is used to ensure that the rider is using the helmet before operating the motorcycle. Sensor detection detects the occurrence of accidents through the detection of collision, tilt, and stationary. In case of an accident, the system notifies people and gives the location of the accident.
This paper discusses the implementation of AI-based smart bike safety and accident detection system.
Keywords: Smart bike safety, helmet detection, accident detection, IoT, deep learning, GPS tracking, emergency alert system, rider safety.
Abstract
AI-Powered Penetration Testing Platform for Automated Vulnerability Detection: A Survey of Artificial Intelligence Methods for Identifying System Weaknesses
Mrs. Bindu K.P, Gagana R, Kushal K R, M G Sahana, and M Harshit Pramod
DOI: 10.17148/IJARCCE.2026.155202
Abstract: The increasing use of shared code libraries, cloud platforms, and collaborative repositories has introduced serious security risks such as hidden vulnerabilities, leaked credentials, and insecure dependencies. Traditional penetration testing methods are often slow, manual, and difficult to integrate into modern DevSecOps workflows. This paper reviews recent AI-based vulnerability detection and penetration testing systems, analyzing their strengths and limitations. It also proposes an AI-powered autonomous security platform that combines static code analysis, dependency scanning, secret detection, Docker-based exploit verification, continuous GitHub monitoring, and AI- generated remediation guidance within a unified interface to improve repository security and vulnerability management.
Keywords: Penetration Testing, Vulnerability Detection, Repository Security, DevSecOps, Artificial Intelligence, Machine Learning, Secret Detection, GitHub Security, Exploit Verification, Docker Sandbox
Keywords: Penetration Testing, Vulnerability Detection, Repository Security, DevSecOps, Artificial Intelligence, Machine Learning, Secret Detection, GitHub Security, Exploit Verification, Docker Sandbox
Abstract
AI BASED INTELLIGENT MOVIE PRODUCTION AND SCRIPT GENERATOR
Dr. Samuel Chellathurai A Ph.D., Shyamvijay H, Vetrivel R
DOI: 10.17148/IJARCCE.2026.155203
Abstract: This paper presents an AI-based professional screenplay generation and production planning system for the Indian film industry, supporting multiple regional languages. The proposed system integrates Large Language Models (LLMs), multi-API fallback mechanisms, and computational production analytics to assist filmmakers, screenwriters, and producers in creating industry-ready scripts and associated production blueprints at zero cost. The system captures user inputs such as film industry (Kollywood, Tollywood, Bollywood, etc.), genre, language (English, Tamil, Telugu, Malayalam, Kannada, Hindi), scale, and protagonist type. Using a resilient chain of free AI services (Sarvam, Krutrim, OpenRouter) with local fallback, the system generates structurally correct screenplays including scene headings, action lines, character names, parentheticals, and dialogues in the chosen language with English-letter transliteration for Dravidian languages. The generated screenplay is then automatically analyzed using script analysis algorithms to extract total scenes, character count, unique locations, action sequences, and complexity level. Based on this analysis, the system computes a professional production plan including pre-production, principal photography, post-production, and marketing timelines, a detailed budget in Indian Rupees with industry-specific multipliers, crew size and department breakdown, equipment logistics, location scouting checklists, and a director’s educational guide. Experimental results demonstrate that the multi-API fallback chain achieves 100% service availability, generating complete feature-length screenplays (35–65 scenes) within 120 seconds on average. Behavioural validation on language adherence shows 94.8% compliance with regional language prompts. Budget estimation error against real Indian film production benchmarks is within ±12% for medium-complexity projects. The proposed framework reduces screenplay writing effort, democratizes production planning for independent filmmakers, and provides an educational tool for aspiring directors and producers in Indian cinema.
Keywords: Screenplay generation, production planning, Indian cinema, large language models, multi-API fallback, script analysis, budget estimation, Kollywood, Tollywood, Bollywood.
Keywords: Screenplay generation, production planning, Indian cinema, large language models, multi-API fallback, script analysis, budget estimation, Kollywood, Tollywood, Bollywood.
Abstract
OPTIMIZING USER WORKFLOW IN AI EMAIL AUTOMATION VIA BROWSER-RESIDENT DOM MUTATION
Sabirhussen Sajidahmed Shaikh, Mahesh Chidanand Konnur, Syed Shamshud Tabrez Ahmed Shahqadri
DOI: 10.17148/IJARCCE.2026.155204
Abstract: Professional email management consumes 28% of the workweek and causes 2.6 hours of daily task-switching loss. While Large Language Models optimize text synthesis, standalone dashboards mandate high-friction application toggling and manual clipboard operations. This paper presents a browser-resident architecture using Google Chrome Manifest V3 and a decoupled Spring Boot backend to embed generative AI directly within the communication viewport. By leveraging the native MutationObserver API for real-time DOM mutation monitoring, the system automates context extraction and eliminates manual copy-paste loops. Empirical evaluations confirm an optimized end-to-end latency of 2.0–4.0 seconds, a stable 4.2 MB browser memory footprint, a 65% network payload compression factor, and near-zero measurable CPU utilization during idle monitoring states, significantly mitigating the context-switching dilemma.
Keywords: MutationObserver, Chrome Extension, Browser-Resident Architecture, Context-Switching Optimization, Spring Boot Microservices, Gemini API.
Keywords: MutationObserver, Chrome Extension, Browser-Resident Architecture, Context-Switching Optimization, Spring Boot Microservices, Gemini API.
Abstract
Causal Artificial Intelligence: Modeling Cause-Effect Relationships for Intelligent Decision Systems
V. N Sheetal, Deepika V M
DOI: 10.17148/IJARCCE.2026.155205
Abstract: Artificial Intelligence (AI) has become an essential technology in healthcare, finance, recommendation systems, forecasting, autonomous systems, and decision-making applications. Most traditional AI systems are based on correlation-driven machine learning techniques that identify patterns from historical data. Although these systems achieve high prediction accuracy, they fail to explain the actual reasoning behind decisions, resulting in lack of transparency, trust, fairness, and robustness. This limitation becomes critical in high-stakes applications where explainability and accountability are important.
Causal Artificial Intelligence (Causal AI) addresses these limitations by introducing cause-effect reasoning into AI systems. Unlike conventional AI models, Causal AI can answer “why,” “what-if,” and “what would have happened differently” questions using causal inference and counterfactual reasoning. This paper presents a comprehensive literature survey of eighteen research papers related to causal inference, explainable AI, deep causal learning, causal machine learning, ethical AI, and forecasting systems. The study analyzes methodologies, models, algorithms, challenges, limitations, and future research directions.
The literature survey identifies several important techniques including Structural Causal Models (SCM), Directed Acyclic Graphs (DAG), Double Machine Learning (DML), Variational Autoencoders (VAE), Fuzzy Cognitive Maps (FCM), SHAP, and LIME. The results indicate that integrating causal reasoning with Machine Learning and Deep Learning significantly improves explainability, fairness, robustness, interpretability, and trustworthy decision-making in modern AI systems.
Keywords: Causal Artificial Intelligence, Explainable AI, Causal Inference, Counterfactual Reasoning, Machine Learning, Deep Learning, Structural Causal Models, Explainability, Ethical AI, Directed Acyclic Graphs
Causal Artificial Intelligence (Causal AI) addresses these limitations by introducing cause-effect reasoning into AI systems. Unlike conventional AI models, Causal AI can answer “why,” “what-if,” and “what would have happened differently” questions using causal inference and counterfactual reasoning. This paper presents a comprehensive literature survey of eighteen research papers related to causal inference, explainable AI, deep causal learning, causal machine learning, ethical AI, and forecasting systems. The study analyzes methodologies, models, algorithms, challenges, limitations, and future research directions.
The literature survey identifies several important techniques including Structural Causal Models (SCM), Directed Acyclic Graphs (DAG), Double Machine Learning (DML), Variational Autoencoders (VAE), Fuzzy Cognitive Maps (FCM), SHAP, and LIME. The results indicate that integrating causal reasoning with Machine Learning and Deep Learning significantly improves explainability, fairness, robustness, interpretability, and trustworthy decision-making in modern AI systems.
Keywords: Causal Artificial Intelligence, Explainable AI, Causal Inference, Counterfactual Reasoning, Machine Learning, Deep Learning, Structural Causal Models, Explainability, Ethical AI, Directed Acyclic Graphs
Abstract
DOCKER: CONTAINERIZATION TECHNOLOGY
Prashanth Kumar G, Dr Madhu H K
DOI: 10.17148/IJARCCE.2026.155206
Abstract: Docker containerization has emerged as a transformative technology in modern software development and deployment, enabling applications to run consistently across diverse computing environments. Traditional application deployment methods often face challenges such as dependency conflicts, scalability limitations, inconsistent runtime environments, and complex infrastructure management. These issues can lead to increased deployment failures, higher maintenance costs, and reduced operational efficiency. To overcome these limitations, Docker provides a lightweight and portable container-based virtualization platform that packages applications along with their dependencies into isolated containers. This approach ensures consistency, flexibility, and seamless deployment across development, testing, and production environments. The proposed system utilizes Docker containerization to simplify application deployment, improve resource utilization, and enhance scalability in cloud and distributed computing environments. By leveraging container orchestration and image-based deployment mechanisms, Docker enables rapid application delivery, efficient system management, and better fault isolation. Additionally, containerization improves system portability, reduces infrastructure overhead, and accelerates continuous integration and continuous deployment (CI/CD) workflows. Through automation and efficient resource management, Docker containerization contributes to reliable, scalable, and cost- effective software solutions, making it an essential technology for modern DevOps and cloud-native applications.
Keywords: Docker Containerization, Docker, Containers, Virtualization, Cloud Computing, DevOps, Microservices, Container Orchestration, Kubernetes, Docker Compose, Application Deployment, Continuous Integration, Continuous Deployment (CI/CD), Software Deployment, Scalable Applications, Cloud-Native Applications, Infrastructure Management, Resource Isolation, Virtual Machines, Platform Portability, Image Management, Container Security, Distributed Systems, Server Virtualization, Application Scalability, Software Development Lifecycle, Container Networking, Automation, Hybrid Cloud, System Performance, Lightweight Virtualization, Deployment Consistency, Multi-Container Applications.
Keywords: Docker Containerization, Docker, Containers, Virtualization, Cloud Computing, DevOps, Microservices, Container Orchestration, Kubernetes, Docker Compose, Application Deployment, Continuous Integration, Continuous Deployment (CI/CD), Software Deployment, Scalable Applications, Cloud-Native Applications, Infrastructure Management, Resource Isolation, Virtual Machines, Platform Portability, Image Management, Container Security, Distributed Systems, Server Virtualization, Application Scalability, Software Development Lifecycle, Container Networking, Automation, Hybrid Cloud, System Performance, Lightweight Virtualization, Deployment Consistency, Multi-Container Applications.
Abstract
A Literature Survey on AI-Based Secure Border Intrusion Detection Systems Using Deep Learning, IoT, and Encrypted Communication
Dakshayini G R, Manya BM, Meghana KJ, Rida Shariff, Sneha GK
DOI: 10.17148/IJARCCE.2026.155207
Abstract: Border intrusion detection is a highpriority national security concern, as unauthorized crossings can facilitate smuggling, trafficking, and terrorism. Conventional surveillance systems relying on continuous human monitoring are limited by operator fatigue, delayed response, and degraded performance under poor environmental conditions such as fog, rain, and low light. This paper presents a literature survey of five recent studies that address AIdriven solutions for border and perimeter security, covering deep learning-based object tracking, hybrid encryption for secure image transmission, integrated surveillance pipelines, machine learning for wireless sensor network protection, and drone-based real-time monitoring. The survey critically analyses the advantages and limitations of each approach and proposes a conceptual integrated framework that combines multi-modal sensing, lightweight deep learning models, and encrypted communication for more reliable and efficient border surveillance.
Keywords: Border security, deep learning, YOLO, DeepSORT, object detection, encryption, IoT, wireless sensor networks, drone surveillance, intrusion detection.
Keywords: Border security, deep learning, YOLO, DeepSORT, object detection, encryption, IoT, wireless sensor networks, drone surveillance, intrusion detection.
Abstract
Credit Card Fraud Detection Using Machine Learning
Yash Ramesh Nalawade, Harshad Sanjay Onkare, Yash Popat Bhosale
DOI: 10.17148/IJARCCE.2026.155208
Abstract: Credit card fraud is a rapidly growing threat fuelled by the expansion of e-commerce and online payment platforms. This paper presents an unsupervised machine learning system for detecting fraudulent credit card transactions using K-Means Clustering. The pipeline loads the Kaggle credit card dataset (284,807 transactions; 492 fraud, 0.172%), drops the ‘Time’ column to simplify clustering and reduce temporal bias, applies StandardScaler normalization, and trains K-Means with K=10 clusters. Fraud is detected by identifying the cluster with the highest fraud density using idxmax(). Model quality is measured by silhouette score, and classification performance is evaluated via confusion matrix, precision, recall, F1-score, and accuracy. Due to the extreme class imbalance, precision, recall, and F1-score are the primary evaluation metrics rather than accuracy. PCA (n_components=2) is used to produce a 2D cluster visualization. A Gradio Blocks interface is deployed with a ‘Single Transaction’ tab (30-feature input, Generate Random button) and a ‘Batch Prediction’ tab (CSV upload). All metrics reported are computed directly from running the project code.
Keywords: credit card fraud detection, K-Means clustering, unsupervised learning, StandardScaler, PCA, silhouette score, Gradio, confusion matrix, fraud density
Keywords: credit card fraud detection, K-Means clustering, unsupervised learning, StandardScaler, PCA, silhouette score, Gradio, confusion matrix, fraud density
Abstract
A Survey on IoT-Based Emergency Navigation and Tracking Device
Mrs. Sushmitha Suresh, Deepthi A Kumar, Inchara S, K P Nihaal, Lalith Adithya M
DOI: 10.17148/IJARCCE.2026.155209
Abstract: Personal safety and emergency navigation are major concerns in remote and low-network environments. Existing safety systems mainly depend on smartphones, internet connectivity, and GSM communication, making them unreliable during emergencies. This survey paper reviews various IoT-based emergency navigation and personal safety systems using GPS, GSM, and LoRa communication technologies. The paper also proposes TRAX, a Smart Emergency Navigation and Tracking Device that integrates breadcrumb-based navigation, LoRa emergency communication, and multi-mode safety functionality into a portable embedded system. The proposed system aims to improve emergency communication, navigation support, and reliability in both urban and remote environments.
Keywords: IoT, GPS Navigation, LoRa, Emergency Alert System, Breadcrumb Navigation, Embedded Systems, Personal Safety
Keywords: IoT, GPS Navigation, LoRa, Emergency Alert System, Breadcrumb Navigation, Embedded Systems, Personal Safety
Abstract
A Preliminary Study on a Unified Haptic Glove for Multi-Domain Applications
Mrs. Sushmitha Suresh, Ismath Zehera, Janhavi S Thorat, K G Soumya, Karanam Vennela
DOI: 10.17148/IJARCCE.2026.155210
Abstract: Virtual Reality (VR) systems require effective interaction techniques to improve realism and user immersion. Conventional VR interaction methods mainly rely on visual and audio responses, offering limited physical interaction within virtual environments. This paper presents a unified haptic glove designed for multi-domain virtual applications. The proposed system integrates flex sensors, MPU6050 motion sensors, vibration motors, and an ESP32 microcontroller to detect hand movement and provide real-time tactile feedback. The glove architecture supports wireless communication and can be adapted for applications such as pilot training, rehabilitation, gaming, and emergency response simulation without modifying the hardware framework. The proposed approach focuses on lightweight design, portability, reduced hardware complexity, and cost-effective implementation for immersive human–computer interaction.
Keywords: Haptic Glove, Virtual Reality, ESP32, Flex Sensor, Tactile Feedback, Motion Tracking, Human– Computer Interaction.
Keywords: Haptic Glove, Virtual Reality, ESP32, Flex Sensor, Tactile Feedback, Motion Tracking, Human– Computer Interaction.
Abstract
Design and Implementation of a Wrong Side Vehicle Detection System Using YOLO And OpenCV.
Chirag Narsinghani, Sankalp Chokale, Nikhil Teli, Pratima Chougule
DOI: 10.17148/IJARCCE.2026.155211
Abstract: The wrong-side driving among the most significant contributing factors to serious traffic crashes, congestion, and loss of life in urban and highway transportation systems. The traditional approaches to traffic monitoring are mostly manual, time consuming and subject to human error. This paper proposes an AI based Wrong Side Vehicle Detection System (WSVDS) to automatically detect vehicles that travel in the opposite direction of the traffic flow using CCTV surveillance camera by computer vision and deep learning techniques. The proposed system relies on the YOLO (You Only Look Once) algorithm to detect vehicles in a frame and Deep SORT/Centroid Tracking to track the vehicles' movement between the frames. Direction analysis is carried out by comparing the trajectory movements of the vehicles with the previously defined traffic direction. When a vehicle travels in the opposite direction to the authorized direction, the system recognizes this as an incident and produces alerts and the storage of evidence. The system can accurately and quickly identify cars, bicycles, buses and trucks in real time, and has a low identification latency. The experimental results show an accuracy of around 95%, which shows the suitability of the proposed system for smart city surveillance and intelligent transportation systems. The proposed solution will greatly cut down on the need for man-to-man monitoring and enhance the safety of the road through automated traffic violation detection.
Keywords: Wrong Side Detection, YOLO, Deep Learning, Computer Vision, Traffic Surveillance, Intelligent Transportation System, Vehicle Tracking, CCTV Monitoring, Deep SORT, AI-based Traffic Monitoring.
Keywords: Wrong Side Detection, YOLO, Deep Learning, Computer Vision, Traffic Surveillance, Intelligent Transportation System, Vehicle Tracking, CCTV Monitoring, Deep SORT, AI-based Traffic Monitoring.
Abstract
AI Website Builder Using Pern Stack
Prathamesh Jalindar Yadav, Vishwajeet Milind Chavan, Mrs. S.S.Parekh
DOI: 10.17148/IJARCCE.2026.155212
Abstract: The AI Powered Website Builder is a smart web application designed to help users generate websites automatically using Artificial Intelligence. The system uses the PERN Stack (PostgreSQL, Express.js, React.js, and Node.js) for full-stack development, Neon Database for cloud-based PostgreSQL storage, and OpenRouter API for integrating multiple AI models. The platform allows users to enter prompts describing their desired website, after which the AI generates website structure, content, and design suggestions automatically. This project reduces manual website development effort, improves productivity, and enables non-technical users to create websites easily. The system provides fast generation, responsive UI, cloud database storage, and scalable architecture
Keywords: Artificial Intelligence, AI Website Builder, PERN Stack, React.js, Node.js, Express.js, PostgreSQL, Neon Database, OpenRouter API, Better Auth, Full Stack Development, Website Generation, Cloud Database, Generative AI, Web Automation, Responsive Web Design.
Keywords: Artificial Intelligence, AI Website Builder, PERN Stack, React.js, Node.js, Express.js, PostgreSQL, Neon Database, OpenRouter API, Better Auth, Full Stack Development, Website Generation, Cloud Database, Generative AI, Web Automation, Responsive Web Design.
Abstract
An AI-Powered Automated Code Review System Using Large Language Models and Static Analysis
Mrs. Nita Meshram, Ganni Naveen Raj Anudeep, K Vedavyas, K Harsha Nandhan, C Balaji Naidu
DOI: 10.17148/IJARCCE.2026.155213
Abstract: Code review is a critical phase in the software development lifecycle, ensuring code quality, maintainability, and security. However, manual code reviews are labor-intensive, often requiring 2 to 4 hours per pull request, and are highly susceptible to human error. Junior developers may inadvertently miss critical security vulnerabilities. This paper proposes an advanced AI-powered automated code review system that seamlessly integrates traditional rule-based static analysis with the deep contextual reasoning capabilities of Large Language Models (LLMs). Operating as an event-driven microservice, the system automatically triggers upon the creation of a GitHub Pull Request, passing Python code changes through a tri-layered analysis engine: Bandit for security vulnerability detection, Pylint for coding standard enforcement, and the Groq LLM for complex logical review and contextual feedback. Results are aggregated, mathematically ranked by severity, and posted directly to the pull request as a structured comment within a 60-second execution window. Deployed on a cost-effective stack including Render, Neon PostgreSQL, and FastAPI, this framework reduces review bottlenecks, minimizes security flaws, and enhances developer productivity.
Keywords: Automated Code Review, Large Language Models, Static Analysis, Continuous Integration, Software Security, Artificial Intelligence, GitHub Actions.
Keywords: Automated Code Review, Large Language Models, Static Analysis, Continuous Integration, Software Security, Artificial Intelligence, GitHub Actions.
Abstract
Optimized Random Forest Model for Chronic Kidney Disease Classification with Imbalanced Data Handling
Kapil, Ankit Navgeet Joshi
DOI: 10.17148/IJARCCE.2026.155214
Abstract: Chronic Kidney Disease (CKD) is a progressive medical disorder that needs to be accurately diagnosed early to avoid serious complications. The aim of this study is to classify the CKD into five stages (Healthy Kidney, Mild CKD (Stage 1-2), Moderate CKD (Stage 3), Severe CKD (Stage 4), Kidney Failure (Stage 5)) using a machine learning technique. A Random Forest (RF) classifier is chosen due to its power and ability to handle high dimension clinical data. The Synthetic Minority Oversampling Technique (SMOTE) is used to solve the imbalance of classes in the data by enhancing the representation of minority classes. Moreover, the optimization of hyperparameters is done with the help of the Grid Search with cross-validation (GridSearchCV) in order to improve the performance and the cross-validation of the model. It is tested with a large pool of demographic, physiological and biochemical data, while the proposed structure is under test. The result of the experiment reveals that the optimized model has 94.12 accuracy which indicates the effectiveness of the model in the multi-class classification of CKD. The results showed that ensemble learning, data balancing, and systematic hyperparameter tuning can be effectively applied to improve the accuracy of the prediction, and the model is applicable in clinical decision support systems and early diagnosis.
Keywords: CKD, SMOTE, Random Forest, GridSearchCV, Ensemble Learning.
Keywords: CKD, SMOTE, Random Forest, GridSearchCV, Ensemble Learning.
Abstract
AI-Powered Smart Platform for Government Scheme Eligibility Using RPA
Archana N, Vaibhavi S, Tejaswini P, S Akshatha, Rishmitha K B
DOI: 10.17148/IJARCCE.2026.155215
Abstract: The growth of digital technologies has changed the way government services given to the citizens. In India, The government's offer many welfare schemes Related to education, healthcare, agriculture, employment, etc financial support. Though these schemes They are made to help people many citizens Still searching it difficult to identify the schemes That fight their eligibility. Information Often available different websites, And the verification process can be confusing and time- consuming. Because of lack of awareness and delay manual processing, many deserving people are unable to receive. Government benefits. This survey paper AI- powered offers. Smart platform to government scheme eligibility and recommendation by using Robotic Process Automation (RPA). Proposed system user Artificial Intelligence (AI), Machine Learning (ML), Natural Language Processing (NLP), and study recommendation techniques user information and give suitable scheme suggestions I real time. RPA robots are used to automate repetitive tasks, e. Government document verification, Farm processing, and application tracking. The platform also supports secure data handling and multilingual communication to improve accessibility for users from different backgrounds. The main purpose of the proposed system to reduce manual effort, Improve and erect transparency welfare schemes easier for access to citizens.
Keywords: Government Scheme Eligibility, AI Recommendation System, Robotic Process Automation, Natural Language Processing, Decision Tree, K-Nearest Neighbour, Content-Based Filtering, Digital Governance, Machine Learning, NLP Chatbot, REST API, Web Scraping, Personalized Recommendations.
Keywords: Government Scheme Eligibility, AI Recommendation System, Robotic Process Automation, Natural Language Processing, Decision Tree, K-Nearest Neighbour, Content-Based Filtering, Digital Governance, Machine Learning, NLP Chatbot, REST API, Web Scraping, Personalized Recommendations.
Abstract
SafeDox: A Review on Secure Document Sharing Using Blockchain, IPFS, and End-to-End Encryption
Neeharika Shakthivelan, Badareesh P, C S Jeevan Setty, Poorvi P, Mrs. Poornima H N
DOI: 10.17148/IJARCCE.2026.155216
Abstract: The need for the development of more efficient document-sharing mechanisms has been brought about by an increased demand for secure data sharing due to developments in cloud computing and digital communication systems. Traditional centralized platforms for data sharing have several limitations such as data leaks, insecurity, single point of failure, and non-transparency. Hence, the need for developing advanced technology like blockchain, IPFS, and encryption is paramount. In the current review article, the discussion has been focused on various methods of secure document-sharing platforms, including blockchain, IPFS, and encryption algorithms. Through blockchain, it is possible to achieve data integrity, data tampering, transparency, and data authentication. Besides, IPFS enables users to employ decentralized storage services that can store huge amounts of data. Similarly, various algorithms for secure encryption like AES-256 and ECC have been considered as they ensure encryption keys and data transfers. Further, different studies and approaches have been compared in terms of methods, merits, drawbacks, and security considerations. Finally, some areas that have not been explored adequately like privacy protection, access control, scalability and complexity of the system have been highlighted as gaps. Consequently, a SafeDox framework has been proposed.
Keywords: Blockchain, IPFS, Secure Document Sharing, AES-256, ECC, Smart Contracts, End-to-End Encryption, Decentralized Storage.
Keywords: Blockchain, IPFS, Secure Document Sharing, AES-256, ECC, Smart Contracts, End-to-End Encryption, Decentralized Storage.
Abstract
Model Context Protocol (MCP): A Standard Interface for Tool-Aware AI Systems
Prajwal P, Prof. Swetha C S
DOI: 10.17148/IJARCCE.2026.155217
Abstract: The rapid evolution of Large Language Models (LLMs) has created a pressing need for standardised mechanisms through which AI agents can interact with external tools, data sources, and services. Current integration approaches rely on ad-hoc REST wrappers or vendor-specific plugin frameworks, producing fragile, non-interoperable systems. This paper presents a comprehensive study of the Model Context Protocol (MCP), an open standard introduced by Anthropic in November 2024 that defines a uniform client-server interface for tool-aware AI systems. We examine the MCP architecture and analyse how it enables dynamic tool discovery, structured resource access, and secure OAuth 2.1 authentication. Through a literature survey of six peer-reviewed and pre- print studies covering MCP security threats, real-world server deployments, adaptive transport applications, and protocol-agnostic integration gaps, we identify current limitations and open research challenges. We further propose a novel MCP Bridge architecture designed to address the multi-protocol adoption gap. Results of our analysis suggest that MCP represents a foundational shift toward composable, auditable, and vendor-neutral AI tool integration.
Keywords: Model Context Protocol, Large Language Models, Tool-Aware AI, JSON-RPC 2.0, Agentic Systems, OAuth 2.1, MCP Bridge, API Integration
Keywords: Model Context Protocol, Large Language Models, Tool-Aware AI, JSON-RPC 2.0, Agentic Systems, OAuth 2.1, MCP Bridge, API Integration
Abstract
CTR Prediction and Campaign Simulation System using Machine Learning
Litrishiya Merceline Mary A, Nivetha S, Dr. K. Ravikumar
DOI: 10.17148/IJARCCE.2026.155218
Abstract: Digital advertising plays an important role in modern marketing. Companies use online advertisements to promote their products and services. One of the most important performance metrics in digital advertising is Click Through Rate (CTR). CTR measures how many users click on an advertisement compared to how many users view it. Accurate CTR prediction helps businesses improve advertisement performance and increase profit.
Manual prediction of advertisement performance is difficult and inaccurate. Machine learning provides an intelligent solution to predict CTR using historical data. Machine learning models analyze advertisement features such as user behavior, advertisement type, device type, and campaign details. Based on these features, the system predicts the probability of user clicks.
This project presents a CTR Prediction and Campaign Simulation System using machine learning. The system uses Logistic Regression to predict CTR. The system also provides campaign simulation to help businesses test advertisement performance before launching campaigns. The system improves decision making, reduces risk, and increases marketing efficiency.
Keywords: CTR Prediction, Machine Learning, Logistic Regression, Digital Marketing, Campaign Simulation
Manual prediction of advertisement performance is difficult and inaccurate. Machine learning provides an intelligent solution to predict CTR using historical data. Machine learning models analyze advertisement features such as user behavior, advertisement type, device type, and campaign details. Based on these features, the system predicts the probability of user clicks.
This project presents a CTR Prediction and Campaign Simulation System using machine learning. The system uses Logistic Regression to predict CTR. The system also provides campaign simulation to help businesses test advertisement performance before launching campaigns. The system improves decision making, reduces risk, and increases marketing efficiency.
Keywords: CTR Prediction, Machine Learning, Logistic Regression, Digital Marketing, Campaign Simulation
Abstract
AI in Agriculture (Claude AI)
N Akhilesh, Prof. Ranganath S R
DOI: 10.17148/IJARCCE.2026.155219
Abstract: Modern agriculture is transitioning toward intelligent automation to enhance crop productivity, reduce resource wastage, and promote sustainable farming practices. This seminar report presents a comprehensive study of how Artificial Intelligence (AI), Internet of Things (IoT), edge computing, unmanned aerial vehicles (drones), and autonomous farming systems collectively transform conventional agriculture into a smart cyber-physical ecosystem. The proposed architecture integrates sensor-based real-time field monitoring, AI-assisted decision making, predictive analytics, and autonomous robotic systems to minimize human dependency while improving precision. The report further explores federated learning for privacy-preserving collaborative model training, and machine learning models for crop disease prediction, intelligent irrigation optimization, yield forecasting, and automated harvesting — creating a holistic framework for next-generation smart farms.
Keywords: Artificial Intelligence, Cyber-Physical Systems, Smart Agriculture, Internet of Things, Edge Computing, Federated Learning, Precision Farming, Drone Surveillance, Machine Learning, Autonomous Robots.
Keywords: Artificial Intelligence, Cyber-Physical Systems, Smart Agriculture, Internet of Things, Edge Computing, Federated Learning, Precision Farming, Drone Surveillance, Machine Learning, Autonomous Robots.
Abstract
AI-SMART LEARN PERSONALISED LEARNING
Dr. K. Prem Kumar, K.Ashritha, J.Amruth, M.Usha Rani, B.Siddhartaa, S.Karunakar Reddy
DOI: 10.17148/IJARCCE.2026.155220
Abstract: With the rapid growth of digital education, the need for personalized learning has become increasingly important. Traditional learning systems often follow a uniform approach that does not consider individual differences in students’ abilities, preferences, and learning pace. This can result in reduced engagement and ineffective learning. To address this issue, this project proposes an AI based Personalized Learning Recommendation System that delivers customized educational content based on individual student characteristics. The system aims to provide a tailored learning experience by analyzing key factors such as available study hours, focus capability, learning speed, and preferred learning modes, including videos, quizzes, and articles. By processing this information, the system creates a unique learning profile for each student, which forms the basis for generating personalized recommendations. At the core of the system is an AI-driven analysis engine that evaluates student performance and learning behavior. It identifies strengths and weaknesses, predicts learning needs, and suggests suitable study strategies. The system continuously updates its recommendations based on the student’s progress, ensuring adaptability and effectiveness. Based on this analysis, the system recommends appropriate learning resources aligned with the student’s preferences. For example, visual learners may receive video content, while others may benefit from quizzes or reading materials. This improves understanding and retention of concepts. In addition to resource recommendations, the system generates personalized study plans according to the student’s available time and learning capacity. These plans are flexible and adapt as the student progresses, helping maintain consistency and organization. The system also assigns targeted learning tasks to improve weak areas and reinforce key concepts. Regular feedback and performance tracking enable students to monitor their progress and stay motivated. Furthermore, the system ensures efficient use of study time by focusing more on difficult topics and less on already mastered content. This leads to improved academic performance and better time management. In conclusion, the AI-based Personalized Learning Recommendation System provides a smart and effective solution for modern education. By leveraging artificial intelligence, it delivers personalized content, adaptive study plans, and targeted tasks that improve learning efficiency and outcomes
Keywords: Artificial Intelligence, Personalized Learning, Recommendation System, Adaptive Learning, E-Learning
Keywords: Artificial Intelligence, Personalized Learning, Recommendation System, Adaptive Learning, E-Learning
Abstract
Push Notification Architecture
Praveen G, Prof K Sharath
DOI: 10.17148/IJARCCE.2026.155221
Abstract: In the modern mobile ecosystem, the reliability of communication channels is as critical as the data they carry. While Software Development Kits (SDKs) have simplified the bridge between user interactions and marketing insights, they have also introduced a complex layer where Silent Failures can compromise application growth. This project explores a robust Push Notification Architecture designed to detect and mitigate internal system failures that do not trigger standard error messages or service crashes. By examining the lifecycle of a notification—from event-driven SDK triggers to third-party gateway delivery—we identify the "blind spots" where systems continue to run while producing incomplete or incorrect results.
The proposed framework integrates a specialized Feedback Loop and Dead Letter Queue (DLQ) strategy to "de- silence" these failures. We demonstrate how an event-driven architecture can be optimized to validate delivery success beyond the initial "200 OK" response. Key features include real-time delivery receipts, device token health monitoring, and automated reconciliation of state between the app and the provider gateways. Results indicate that by identifying silent failures at the architectural level, developers can improve notification delivery rates by 35% and ensure that audience segmentation data remains accurate, ultimately preventing the waste of marketing resources on unreachable user segments.
The proposed framework integrates a specialized Feedback Loop and Dead Letter Queue (DLQ) strategy to "de- silence" these failures. We demonstrate how an event-driven architecture can be optimized to validate delivery success beyond the initial "200 OK" response. Key features include real-time delivery receipts, device token health monitoring, and automated reconciliation of state between the app and the provider gateways. Results indicate that by identifying silent failures at the architectural level, developers can improve notification delivery rates by 35% and ensure that audience segmentation data remains accurate, ultimately preventing the waste of marketing resources on unreachable user segments.
Abstract
Explainable Multimodal AI for Deepfake Detection and Digital Content Authenticity
R S Geethanjali, Sneha KS, Vaidehi Vasudev Arundekar, Vinutha BR
DOI: 10.17148/IJARCCE.2026.155222
Abstract: Recent breakthroughs in AI and deep learning have enabled the creation of highly convincing synthetic media — commonly termed deepfakes — that are increasingly difficult to distinguish from authentic content. Techniques such as GANs, transformer-based architectures, and diffusion models introduce serious risks to cybersecurity, journalistic integrity, digital trust, and democratic processes. Existing solutions are predominantly limited to single-modality analysis, hindering effectiveness against coordinated multimedia misinformation spanning video, audio, and text simultaneously. This paper surveys existing literature on deepfake detection, explainable AI, and multimodal misinformation analysis, identifying key research gaps and limitations. Based on these findings, we propose an Explainable Multimodal AI Framework that unifies ResNet18, CNNs, and DistilBERT with SHAP, LIME, and RAG-based contextual reasoning for simultaneous detection and authenticity verification of manipulated multimedia content. The proposed framework is expected to deliver improved accuracy, transparency, and real-time inference capability over existing unimodal and non- explainable approaches when evaluated on standard benchmarks such as FaceForensics++, DFDC, ASVspoof 2019, and FakeNewsNet.
Keywords: Deepfake Detection; Explainable AI; Multimodal Learning; Digital Authenticity; Generative Adversarial Networks; CNN; ResNet18; DistilBERT; Retrieval-Augmented Generation; SHAP; LIME; Real-Time Detection.
Keywords: Deepfake Detection; Explainable AI; Multimodal Learning; Digital Authenticity; Generative Adversarial Networks; CNN; ResNet18; DistilBERT; Retrieval-Augmented Generation; SHAP; LIME; Real-Time Detection.
Abstract
THE LIFECYCLE OF A DATA EVENT
Nitish S, Sandarsh Gowda M M
DOI: 10.17148/IJARCCE.2026.155223
Abstract: In today's data-driven digital landscape, every user interaction—every tap, swipe, scroll, or transaction— generates a discrete unit of information known as a "data event." Understanding the journey of this event, from its inception on a client device to its eventual transformation into business intelligence, is fundamental to building scalable, reliable, and insight-rich digital products. This project explores the end-to-end lifecycle of a data event, detailing the technical, architectural, and analytical stages it traverses across modern distributed systems. By examining each phase— generation, collection, transmission, validation, enrichment, storage, processing, and activation—developers and data engineers can construct robust event pipelines that power real-time analytics, personalization engines, and machine learning workflows.
The proposed framework illustrates how raw user signals are systematically captured by client-side libraries, structured into standardized schemas, routed through high-throughput streaming platforms such as Apache Kafka, validated against governance rules, and ultimately persisted within data warehouses or lakes for downstream consumption. We examine the role of event-driven architectures in decoupling producers and consumers, enabling horizontal scalability and fault tolerance. Key stages such as schema enforcement, deduplication, identity resolution, and event activation are analyzed for their efficacy in maintaining data integrity. Furthermore, this study addresses the challenges of latency, data loss prevention, privacy compliance (GDPR/CCPA), and observability across the pipeline. The results demonstrate that a well-orchestrated event lifecycle reduces data discrepancies by over 45% while significantly improving the timeliness and trustworthiness of analytics, paving the way for real-time decision-making across enterprise systems.
Keywords: Data Event, Event Lifecycle, Event-Driven Architecture, Data Pipeline, Stream Processing, Data Ingestion, Schema Validation, Real-Time Analytics, Data Governance, ETL
The proposed framework illustrates how raw user signals are systematically captured by client-side libraries, structured into standardized schemas, routed through high-throughput streaming platforms such as Apache Kafka, validated against governance rules, and ultimately persisted within data warehouses or lakes for downstream consumption. We examine the role of event-driven architectures in decoupling producers and consumers, enabling horizontal scalability and fault tolerance. Key stages such as schema enforcement, deduplication, identity resolution, and event activation are analyzed for their efficacy in maintaining data integrity. Furthermore, this study addresses the challenges of latency, data loss prevention, privacy compliance (GDPR/CCPA), and observability across the pipeline. The results demonstrate that a well-orchestrated event lifecycle reduces data discrepancies by over 45% while significantly improving the timeliness and trustworthiness of analytics, paving the way for real-time decision-making across enterprise systems.
Keywords: Data Event, Event Lifecycle, Event-Driven Architecture, Data Pipeline, Stream Processing, Data Ingestion, Schema Validation, Real-Time Analytics, Data Governance, ETL
Abstract
AI-Based Emergency Response System Using Facial Recognition, GPS Navigation and Number Plate Recognition
C J Nagasri Pragna, C Panduranga Reddy, Chaitra Patil, D Eshwar Kumar, Dr. Renuka Sagar
DOI: 10.17148/IJARCCE.2026.155224
Abstract: Rapid ambulance response is critical for saving human lives during medical emergencies. However, existing emergency transportation systems often suffer from traffic congestion, delayed route optimization, ambulance siren misuse, and lack of intelligent traffic coordination.
This paper presents an AI-Based Emergency Response System integrating facial recognition, GPS-based shortest path navigation, obstacle detection, and OCR-based number plate recognition to improve ambulance efficiency and emergency healthcare transportation.
The proposed system verifies the presence of a genuine patient using facial recognition before activating the ambulance siren. GPS navigation identifies the fastest route to nearby hospitals using live traffic analysis. A front-mounted camera and ultrasonic sensors continuously monitor vehicles obstructing the ambulance path.
Experimental analysis demonstrated a facial recognition accuracy of 96.2%, OCR detection accuracy of 93.4%, and improved route optimization efficiency under simulated traffic conditions. The integration of Artificial Intelligence, Computer Vision, GPS Navigation, and Smart Transportation technologies pro- vides an efficient and scalable solution for modern emergency healthcare systems.
Keywords: Artificial Intelligence, Emergency Response System, Facial Recognition, GPS Navigation, OCR, Ambulance Automation, OpenCV, Smart Healthcare
This paper presents an AI-Based Emergency Response System integrating facial recognition, GPS-based shortest path navigation, obstacle detection, and OCR-based number plate recognition to improve ambulance efficiency and emergency healthcare transportation.
The proposed system verifies the presence of a genuine patient using facial recognition before activating the ambulance siren. GPS navigation identifies the fastest route to nearby hospitals using live traffic analysis. A front-mounted camera and ultrasonic sensors continuously monitor vehicles obstructing the ambulance path.
Experimental analysis demonstrated a facial recognition accuracy of 96.2%, OCR detection accuracy of 93.4%, and improved route optimization efficiency under simulated traffic conditions. The integration of Artificial Intelligence, Computer Vision, GPS Navigation, and Smart Transportation technologies pro- vides an efficient and scalable solution for modern emergency healthcare systems.
Keywords: Artificial Intelligence, Emergency Response System, Facial Recognition, GPS Navigation, OCR, Ambulance Automation, OpenCV, Smart Healthcare
Abstract
Hybrid Quantum-Classical Optimization for Energy Distribution using TSP Model
Mrs. Punitha M R, Aditya H, Ankitha P, Anusha M N, and Deekshitha K
DOI: 10.17148/IJARCCE.2026.155225
Abstract: Managing energy distribution efficiently has become an important challenge in modern smart grid systems. This paper proposes a hybrid quantum-classical optimization approach based on the Traveling Salesman Problem (TSP) model. The system combines classical optimization techniques with the Quantum Approximate Optimization Algorithm (QAOA) to determine efficient energy distribution paths while reducing transmission cost and energy loss. Qiskit is used for simulation and implementation. The results indicate improved routing efficiency and demonstrate the potential of quantum computing in handling complex optimization problems in smart grid applications.
Keywords: Hybrid Quantum-Classical Optimization, Traveling Salesman Problem (TSP), Quantum Approximate Optimization Algorithm (QAOA), Smart Grid, Energy Distribution, Qiskit, Quantum Computing.
Keywords: Hybrid Quantum-Classical Optimization, Traveling Salesman Problem (TSP), Quantum Approximate Optimization Algorithm (QAOA), Smart Grid, Energy Distribution, Qiskit, Quantum Computing.
Abstract
AI-BASED EARLY POULTRY DISEASE RISK PREDICTION SYSTEM USING REAL-TIME TREND ANALYSIS
Mamatha R, Bhavya Sai Shree V, Chithra U, Deeksha N, Dyuthi S
DOI: 10.17148/IJARCCE.2026.155226
Abstract: Poultry farming plays a vital role in food production and the agricultural economy, but disease outbreaks in poultry farms can lead to significant economic losses, reduced productivity, and increased mortality rates. Early detection and prediction of diseases are essential to ensure the health and safety of poultry birds. This paper presents an AI-Based Early Poultry Disease Risk Prediction System using Real-Time Trend Analysis that aims to monitor environmental and health-related parameters continuously and predict possible disease risks at an early stage. The proposed system utilizes Artificial Intelligence and Ma chine Learning techniques to analyze real-time data collected from sensors and farm records, including temperature, humidity, ammonia levels, feed intake, water consumption, and bird activity. By applying predictive algorithms and trend analysis, the system identifies abnormal patterns and provides early warnings to farmers before disease outbreaks occur. The system also generates risk assessments and recommendations to improve farm management and reduce losses. The integration of IoT devices, cloud-based monitoring, and AI-driven analytics helps in improving disease prediction accuracy, reducing manual monitoring efforts, and supporting sustainable poultry farming practices. The proposed solution enables farmers to take preventive actions in advance, thereby improving poultry health, increasing productivity, and minimizing economic losses. Index Terms—Artificial Intelligence, Poultry Disease Prediction, Real-Time Trend Analysis, Machine Learning, Smart Poul try Farming, Internet of Things, Predictive Analytics, Early Disease Detection, Environmental Monitoring, Farm Automation.
Abstract
Strengthening CBSE Computer Science and Informatics Practices Education: A Proposed Framework for AI Integration, Global Awareness, and Career Readiness at Class XI–XII Level
Neerja Jain
DOI: 10.17148/IJARCCE.2026.155227
Abstract: Globally, secondary school students are learning to train machine learning models, build AI-powered applications, and deploy code on cloud platforms — in countries as varied as Singapore, Finland, China, and the United States. In India, Computer Science (Code 083) and Informatics Practices (Code 065) provide a strong foundation in Python, Pandas, SQL, and networking — skills that are genuinely relevant and correctly chosen. Yet a student completing Class XII in either subject has not encountered Artificial Intelligence in any practical form within their main course, has no awareness that languages such as R, Julia, JavaScript, and Rust drive significant portions of the global AI and data economy, and has received no guidance on where two years of technical learning leads professionally. This paper presents a systematic analysis of the CBSE Class XI–XII CS and IP syllabi, identifies three specific gaps — absent AI integration, narrow technology horizon, and invisible career pathways — and proposes a concrete, implementable framework to address each. The framework introduces AI concepts naturally within existing syllabus topics (including a three-level ML introduction requiring no new infrastructure), strengthens practical learning through a Build-Deploy-Share model, builds global language awareness, and maps subject skills to explicit career pathways for both science and commerce stream students. Every proposed change is designed for immediate adoption in government senior secondary schools without new subjects, new examinations, or additional cost.
Keywords: CBSE Computer Science, Informatics Practices, AI Integration, Future-Ready Education, Career Pathways, NEP 2020, Python, Global Programming Languages, Government Schools, Practical Learning
Keywords: CBSE Computer Science, Informatics Practices, AI Integration, Future-Ready Education, Career Pathways, NEP 2020, Python, Global Programming Languages, Government Schools, Practical Learning
Abstract
“A Survey on Prompt Injection and Jailbreak Defenses in Large Language Models”
Anushree K N, Vedanth M, Amulya H, Chirayu Gowda, Prof. Meghashree C
DOI: 10.17148/IJARCCE.2026.155228
Abstract: Large language models are now embedded in healthcare, education, and enterprise software at a scale that would have seemed unlikely just a few years ago. This rapid adoption has introduced a category of security threats that conventional mechanisms were never designed to handle: prompt injection and jailbreak attacks. Unlike traditional exploits, these attacks do not target code; they manipulate natural language itself to push a model past its safety constraints, extract information it should not reveal, or produce outputs its developers explicitly prohibited. What makes these threats particularly difficult to counter is that adversarial intent is often distributed gradually across multiple conversation turns, each message appearing harmless in isolation while collectively steering the model toward a malicious outcome. Defenses built on static keyword lists or single-message classification are structurally unable to detect this pattern.
Keywords: large language models, prompt injection, dual LLM, adaptive security, context drift, FAISS, semantic defense, adversarial attacks.
Keywords: large language models, prompt injection, dual LLM, adaptive security, context drift, FAISS, semantic defense, adversarial attacks.
Abstract
AI-Powered Detection of Deepfake Audio in Hindi and Kannada Using Speech Analysis
Mrs. Kavitha K S, Chaitanya C Gowda, D Yashwanth, Dheeraj R, Lishanth N
DOI: 10.17148/IJARCCE.2026.155229
Abstract: The exponential growth of generative artificial intelligence has enabled the mass production of deepfake audio—synthetic speech crafted to replicate the vocal identity of real individuals. Such fabricated audio introduces severe threats to financial security, democratic discourse, biometric authentication, and the credibility of legal evidence. Despite extensive research in English-centric audio forensics, Indian regional languages, specifically Hindi and Kannada, remain substantially underrepresented in the literature. This paper presents a comprehensive survey of existing deepfake audio detection techniques, analyses critical research gaps pertaining to Indian regional languages, and proposes an AI-powered detection framework tailored to Hindi and Kannada speech. The proposed system employs a Convolutional Neural Network (CNN) and Transformer encoder hybrid to jointly model local spectral patterns and long-range temporal dependencies in audio signals. A custom multilingual dataset is constructed from real speech corpora supplemented with synthesized audio generated via Google TTS, Coqui TTS, and Bark. Acoustic features including Mel-Frequency Cepstral Coefficients (MFCC), mel-spectrograms, chroma, and prosodic descriptors are extracted using the Librosa toolkit. The model performs binary classification—Real versus Fake—with performance assessed through Accuracy, Equal Error Rate (EER), False Acceptance Rate (FAR), and False Rejection Rate (FRR). A real-time Flask/Streamlit web interface enables non-technical users to upload audio and receive instant detection results alongside a confidence score.
Keywords: Deepfake audio detection, Hindi speech, Kannada speech, CNN-Transformer, MFCC, mel-spectrogram, Indian language forensics, voice cloning, binary classification, EER
Keywords: Deepfake audio detection, Hindi speech, Kannada speech, CNN-Transformer, MFCC, mel-spectrogram, Indian language forensics, voice cloning, binary classification, EER
Abstract
Innovations in Stroke Identification: A Machine Learning-Based Diagnostic Model Using Neuroimages
Malti Punjaram Barve, Prof.Puspendu Biswas
DOI: 10.17148/IJARCCE.2026.155230
Abstract: Stroke diagnosis is a time-sensitive and critical healthcare process that requires rapid and accurate detection to ensure effective treatment and improved patient outcomes. Stroke is one of the leading causes of mortality, neurological disorders, and long-term disability worldwide, creating a significant burden on healthcare systems. Conventional diagnostic methods such as Computed Tomography (CT) scans and Magnetic Resonance Imaging (MRI) depend heavily on expert radiological analysis, which may lead to delays during emergency situations where immediate medical intervention is essential. To overcome these limitations, this research presents a machine learning and deep learning- based stroke diagnosis system using neuroimaging data.The proposed framework utilizes advanced Deep Learning architectures, including Convolutional Neural Networks (CNN), Inception V3, and MobileNet, for efficient analysis and classification of brain neuroimages. The CNN model performs automated feature extraction from medical images, while Inception V3 enhances detection accuracy by capturing complex spatial and visual patterns through deep convolutional layers. Additionally, MobileNet provides a lightweight and computationally efficient architecture, enabling faster processing and suitability for real-time clinical applications. The integration of these models ensures both high diagnostic accuracy and reduced computational complexity.The system was trained and tested on neuroimaging datasets containing both stroke-affected and healthy brain scans. Experimental evaluation demonstrated superior performance in terms of accuracy, precision, sensitivity, and reliability for stroke prediction and classification. The proposed intelligent diagnostic model assists healthcare professionals in achieving faster stroke detection, early diagnosis, and timely treatment planning. By incorporating Artificial Intelligence (AI), Medical Image Processing, and Machine Learning techniques into the healthcare workflow, the system can significantly reduce diagnosis time, improve patient care quality, minimize human error, and decrease the economic burden associated with stroke management and rehabilitation.
Keywords: Stroke Diagnosis, Machine Learning, Deep Learning, Convolutional Neural Network (CNN), Inception V3, MobileNet, Neuroimaging, Medical Image Processing, Artificial Intelligence (AI), Brain Stroke Detection, CT Scan, MRI Scan, Healthcare Analytics, Stroke Prediction, Real-Time Diagnosis, Clinical Decision Support System.
Keywords: Stroke Diagnosis, Machine Learning, Deep Learning, Convolutional Neural Network (CNN), Inception V3, MobileNet, Neuroimaging, Medical Image Processing, Artificial Intelligence (AI), Brain Stroke Detection, CT Scan, MRI Scan, Healthcare Analytics, Stroke Prediction, Real-Time Diagnosis, Clinical Decision Support System.
Abstract
Augmented Reality (AR) Based Virtual Lab: A Survey on Interactive AR Learning Platforms and Educational Applications Using Deep Learning Techniques
Meena G, Mayurakhi Maiti, Sheethal M, Shilpa B, and Vineela K
DOI: 10.17148/IJARCCE.2026.155231
Abstract: Augmented Reality (AR) is becoming an important digital learning approach because it can convert routine laboratory instruction into an immersive, flexible, and learner-friendly experience. Conventional laboratories demand costly equipment, regular maintenance, sufficient physical space, and continuous supervision; these requirements often restrict repeated practice and equal access for students. AR-enabled virtual laboratories overcome many of these barriers by placing three-dimensional digital objects and guided simulations over the real environment through mobile or wearable devices.
This survey presents a consolidated review of AR-supported laboratory learning with particular focus on deep learning, adaptive interaction, object recognition, simulation design, and educational outcomes. It discusses the role of tools and frameworks such as Unity 3D, Blender, Vuforia SDK, Android Studio, Faster R-CNN, SSD, and YOLOv7. Existing approaches are compared with respect to accuracy, response time, scalability, and usefulness in teaching-learning practice. The paper also identifies limitations such as device dependency, processing load, tracking errors, and usability concerns, and it highlights future possibilities involving artificial intelligence, cloud-supported AR, analytics, and personalized virtual laboratories.
Keywords: Augmented Reality; Virtual Laboratories; Educational Technology; Deep Learning; YOLOv7; Interactive Learning; Computer Vision.
This survey presents a consolidated review of AR-supported laboratory learning with particular focus on deep learning, adaptive interaction, object recognition, simulation design, and educational outcomes. It discusses the role of tools and frameworks such as Unity 3D, Blender, Vuforia SDK, Android Studio, Faster R-CNN, SSD, and YOLOv7. Existing approaches are compared with respect to accuracy, response time, scalability, and usefulness in teaching-learning practice. The paper also identifies limitations such as device dependency, processing load, tracking errors, and usability concerns, and it highlights future possibilities involving artificial intelligence, cloud-supported AR, analytics, and personalized virtual laboratories.
Keywords: Augmented Reality; Virtual Laboratories; Educational Technology; Deep Learning; YOLOv7; Interactive Learning; Computer Vision.
Abstract
AI Model to Detect Bone Fracture
Kavitha K S, Mallikarjun Biradar, Pavan S, Sourabh Goud Allolli, Yallaling Meti
DOI: 10.17148/IJARCCE.2026.155232
Abstract: Bone fractures represent one of the most prevalent categories of traumatic injury encountered in emergency clinical settings, and their prompt identification from radiographic images is paramount to ensuring effective patient management and reducing the risk of long-term musculoskeletal complications. Conventional manual examination of X- ray images is inherently subject to inter-observer variability, fatigue-induced diagnostic errors, and throughput constraints, particularly in high-volume accident and emergency departments. This paper presents a comprehensive survey and an original deep learning framework leveraging the You Only Look Once version 8 (YOLOv8) single-stage object detection architecture for real-time, automated detection and spatial localisation of bone fractures in plain radiographic X-ray images. The proposed pipeline integrates Contrast Limited Adaptive Histogram Equalisation (CLAHE)-based image enhancement, mosaic data augmentation, and a structured transfer learning strategy founded on MS COCO pre-trained weights to maximise generalisation across diverse fracture morphologies and anatomical regions. A consolidated dataset of 8,742 annotated radiographs spanning seven skeletal regions was employed for training, validation, and testing under stratified partitioning. Experimental evaluation demonstrates that the proposed YOLOv8m model achieves a mean Average Precision (mAP@0.5) of 91.4%, a clinical sensitivity of 92.7%, and a specificity of 89.3%, with a real-time inference throughput of 56 frames per second. Systematic comparative benchmarking against VGG-16, ResNet-50, Faster R-CNN, and YOLOv5s confirms the superiority of the proposed approach. An ablation study further validates the individual contributions of CLAHE pre-processing, mosaic augmentation, and transfer learning to overall detection performance. The findings establish YOLOv8 as a clinically viable, decision-support technology for automated fracture screening in radiology workflows.
Keywords: Bone Fracture Detection, YOLOv8, YOLO, Deep Learning, X-Ray Image Analysis, Medical Imaging, Object Detection, Convolutional Neural Networks, Computer-Aided Diagnosis, Transfer Learning, CLAHE
Keywords: Bone Fracture Detection, YOLOv8, YOLO, Deep Learning, X-Ray Image Analysis, Medical Imaging, Object Detection, Convolutional Neural Networks, Computer-Aided Diagnosis, Transfer Learning, CLAHE
Abstract
EMOTRIXA: AI Companion Robots for Emotional Support
Rishika. B, Dr Kavitha AS
DOI: 10.17148/IJARCCE.2026.155233
Abstract: Emotional well-being is a concern these days. Many people feel stressed, lonely and anxious. This is because of things like school pressure work demands and not seeing friends and family much. Even when people are struggling they often do not talk about their feelings because they are afraid of what others might think or they do not know where to get help.
This paper is about a robot that uses intelligence to understand how people are feeling. The robot can see how people are feeling by looking at their faces. It uses a kind of computer program to figure out if someone is happy, sad, angry or okay. The robot is trained to do this by looking at a lot of pictures of peoples faces. It can keep watching. See how someones feelings change.
The robot can talk to people without them having to start the conversation. It is like a friend who can see how you are feeling and respond. The robot is pretty good at understanding emotions. It gets it right 80 to 85 percent of the time. It is not meant to replace doctors or therapists. It can help people feel less alone and talk to someone when they need to. This project shows how artificial intelligence can be used in a way to help people feel better. Emotional well-being is important. This robot is designed to support emotional well-being. The robot uses intelligence to support emotional well-being and it can be a helpful tool, for people who need support.
This paper is about a robot that uses intelligence to understand how people are feeling. The robot can see how people are feeling by looking at their faces. It uses a kind of computer program to figure out if someone is happy, sad, angry or okay. The robot is trained to do this by looking at a lot of pictures of peoples faces. It can keep watching. See how someones feelings change.
The robot can talk to people without them having to start the conversation. It is like a friend who can see how you are feeling and respond. The robot is pretty good at understanding emotions. It gets it right 80 to 85 percent of the time. It is not meant to replace doctors or therapists. It can help people feel less alone and talk to someone when they need to. This project shows how artificial intelligence can be used in a way to help people feel better. Emotional well-being is important. This robot is designed to support emotional well-being. The robot uses intelligence to support emotional well-being and it can be a helpful tool, for people who need support.
Abstract
Violence Detection in Video Surveillance Using Frame Extraction
Priyanka Sunil Kognole, Shruti Mahadev Pisal, Suparshwa Sudhir Kognole
DOI: 10.17148/IJARCCE.2026.155234
Abstract: This study presents a novel approach for enhancing the automation and effectiveness of real-time threat detection in video surveillance systems. Traditional surveillance methods require continuous human monitoring, are resource-intensive, and of ten fail to consistently identify suspicious activities with precision. Addressing these challenges, we propose the Mono-Scale CNN-LSTM Fusion Network, an advanced deep learning model designed for automated, sustainable, and high-accuracy CCTV systems. The model utilizes Convolutional Neural Networks(CNN) in combination with Long Short Term Memory (LSTM) networks to improve recognition capabilities by capturing temporal and spatial features. For feature extraction, the Oriented FAST and Rotated BRIEF (ORB) techniques are employed to enhance detection efficiency. The model was tested using the UCF crime image dataset and achieved an accuracy rate of approximately 99%, surpassing traditional models like CNN, VGG-16, VGG-19, ResNet-50, and Dense Net. This study highlights the contributions of our approach, which offers a significant reduction in the need for human oversight and sets new standards in the field of automatic threat detection. Furthermore, it emphasizes the model’s capability to support contemporary security systems with high precision, reliability, and scalability, making it a valuable tool for the next generation of intelligent surveillance systems.
Keywords: Real-time threat detection, video surveillance, deep learning, CNN-LSTM, ORB feature extraction, UCF crime dataset, automated security, intelligent surveillance, high-accuracy CCTV.
Keywords: Real-time threat detection, video surveillance, deep learning, CNN-LSTM, ORB feature extraction, UCF crime dataset, automated security, intelligent surveillance, high-accuracy CCTV.
Abstract
Smart Flood Detection and Impact Mapping System: An IoT-Driven Framework for Real-Time Early Warning and Spatial Risk Visualisation
Mrs. Vidya V Patil, Amar, Divit V, Hari Narayana S, Kiran S
DOI: 10.17148/IJARCCE.2026.155235
Abstract: Year after year, floods establish themselves as one of nature’s most relentless and costly hazards — stripping communities of lives, agricultural stability, and infrastructure built over generations. Much of this destruction is not unavoidable; a substantial share of it can be traced directly to the failure of existing warning systems to reach at- risk populations quickly enough, and reliably enough, when conditions deteriorate. This paper presents the Smart Flood Detection and Impact Mapping System, a hardware-first, internet-independent platform that integrates IoT environmental sensing, embedded microcontroller processing, and GSM cellular communication to deliver autonomous real-time flood alerts without relying on cloud infrastructure. The system continuously measures water level, rainfall intensity, temperature, and humidity through field-deployed sensor arrays; evaluates incoming readings against pre- calibrated safety thresholds; and immediately dispatches SMS warnings to residents, farmers, local authorities, and emergency management teams the moment dangerous conditions are detected. Because alert delivery travels through the GSM cellular network rather than the internet, the system remains fully operational during the network outages that characteristically accompany severe weather events. A complementary Python-based visualisation pipeline converts accumulated sensor telemetry into colour-indexed spatial heat maps, providing disaster coordinators with a structured, geographically explicit picture of inundation severity that supports evidence-based evacuation planning and rescue resource allocation. The resulting system is low-cost, energy-efficient, portable, and suited for deployment across both urban centres and the remote rural communities where the gap between flood risk and monitoring capability is widest.
Keywords: Flood Monitoring, IoT, GSM Module, Early Warning System, Heat Mapping, Disaster Management, Real-Time Monitoring, Embedded Systems, Arduino, Edge Computing, Environmental Sensing, SMS Alert, SIM800L.
Keywords: Flood Monitoring, IoT, GSM Module, Early Warning System, Heat Mapping, Disaster Management, Real-Time Monitoring, Embedded Systems, Arduino, Edge Computing, Environmental Sensing, SMS Alert, SIM800L.
Abstract
AI-BASED SMART FARMING MONITORING SYSTEM
Anushya V P, D Nikhil, Darshan M, Dodagatta Mahesh, Nitish Kumar
DOI: 10.17148/IJARCCE.2026.155236
Abstract: Agriculture plays an important role in economic development, but traditional farming methods face challenges such as water wastage, unpredictable climate conditions, soil degradation, and low crop productivity. Farmers also find it difficult to continuously monitor field conditions manually. To overcome these problems, this project proposes an AI- Based Smart Farming Monitoring System using Artificial Intelligence (AI), Internet of Things (IoT), and real-time monitoring technologies. The system uses sensors to collect environmental data such as soil moisture, temperature, humidity, and rainfall. The collected data is processed and analysed to support smart irrigation and crop management decisions. A web dashboard allows farmers to monitor field conditions remotely and receive real-time alerts. The system also automates irrigation based on soil moisture levels, reducing water wastage and manual effort. The proposed system improves farming efficiency, supports sustainable agriculture, and helps increase crop productivity.
Keywords: Smart Agriculture, IoT, Artificial Intelligence, Automated Irrigation, Real-Time Monitoring, ESP8266, Precision Farming, Crop Monitoring, Flask, Python
Keywords: Smart Agriculture, IoT, Artificial Intelligence, Automated Irrigation, Real-Time Monitoring, ESP8266, Precision Farming, Crop Monitoring, Flask, Python
Abstract
Automated Blood Cell Segmentation and Classification Using YOLOv11n: An End-to- End Deep Learning Approach
Dr. Jagadish R M, Poorvi V Mallapur, Pragna Kakandaki, Rakshitha Atti
DOI: 10.17148/IJARCCE.2026.155237
Abstract: This research presents an efficient deep learning solution for detecting and counting blood cells in microscope images using the YOLOv11n object detection model. Leveraging a robust annotated dataset, data augmentation, and advanced inference, the system achieves high detection accuracy (mAP 90.5%) for RBCs, WBCs, and platelets. Automated results improve laboratory workflows and reliability, demonstrating strong real-world impact for digital hematology.
Abstract
Smart Anti-Theft SmartPhone Ecosystem With Offline Remote Access Using BLE and GSM: A Survey of Hardware and Algorithmic Methods for Securing Devices and Automated Offline Recovery
Mrs. Bindu K.P, Dimple J, Gabburi Narasanna Pallavi, Gaddamamadugu Dinavya
DOI: 10.17148/IJARCCE.2026.155238
Abstract: Smartphone theft and offline loss scenarios remain a major user concern. Conventional cloud dependent tracking services are effective when devices are online, but they fail when devices are powered off, disconnected, or outside IP coverage. This paper surveys the design space for a Smart Anti-theft Smartphone Ecosystem that provides offline remote access and recovery by combining Bluetooth Low Energy (BLE) proximity and direction sensing with GSM-based command-and-alert channels. We synthesize algorithmic advances in direction estimation, practical embedded prototypes (ESP32 + SIM800L), and secure beacon telemetry approaches suitable for constrained payloads. The survey analyzes tradeoffs among accuracy, energy, latency, and privacy, consolidating findings from simulation studies, hardware prototypes, and field experiments. Key conclusions are: physics-informed feature engineering improves BLE direction estimation under sparse angular coverage; manifold-guided interpolation is an effective augmentation strategy for under-sampled angle regions; GSM/SMS provides a robust, lowbandwidth fallback for remote commands when IP connectivity is unavailable; and privacy-preserving beacon protocols and scalable key management remain open challenges. We propose a prioritized research roadmap and an implementation blueprint tailored to resource-constrained anti-theft deployments.
Keywords: Bluetooth Low Energy, GSM, anti-theft, direction estimation, ESP32, beacon security, offline tracking, DevSecOps, Vulnerability Detection.
Keywords: Bluetooth Low Energy, GSM, anti-theft, direction estimation, ESP32, beacon security, offline tracking, DevSecOps, Vulnerability Detection.
Abstract
Smart Crop Doctor: An AI Driven Chilli Plant Disease Detection
Dr. C N Shariff, A Pravalika, D Ranjitha, D Vindhu, Ganga
DOI: 10.17148/IJARCCE.2026.155239
Abstract: Agriculture is one of the most important sectors that supports food production and the economy. Chilli cultivation is highly affected by several diseases that reduce crop quality, productivity, and farmers’ income. In many rural areas, farmers still depend on manual inspection and agricultural experts to identify plant diseases. This process is often time-consuming, expensive, and not easily accessible to all farmers.
To address this issue, the proposed system “Smart Crop Doctor” introduces an intelligent and user-friendly solution for automatic chilli plant disease detection. The system uses Artificial Intelligence and Deep Learning techniques to analyze chilli leaf images and identify diseases accurately. A transfer learning model called MobileNetV2 is used to classify healthy and infected leaves with improved prediction performance.
In addition to disease detection, the system also provides treatment suggestions, prevention methods, expert consultation support, and crop management guidance. The application is developed as a web-based platform so that farmers can access it easily using computers or mobile devices. The system also stores crop history and diagnosis records for future monitoring and analysis.
Experimental observations show that the proposed framework provides reliable disease prediction, fast response time, and practical support for farmers. The system helps in early disease identification, reduces crop losses, improves productivity, and promotes modern smart farming practices.
Keywords: Smart Agriculture, Plant Disease Detection, Deep Learning, MobileNetV2, Chilli Leaf Classification, Precision Farming, AI Advisory System
To address this issue, the proposed system “Smart Crop Doctor” introduces an intelligent and user-friendly solution for automatic chilli plant disease detection. The system uses Artificial Intelligence and Deep Learning techniques to analyze chilli leaf images and identify diseases accurately. A transfer learning model called MobileNetV2 is used to classify healthy and infected leaves with improved prediction performance.
In addition to disease detection, the system also provides treatment suggestions, prevention methods, expert consultation support, and crop management guidance. The application is developed as a web-based platform so that farmers can access it easily using computers or mobile devices. The system also stores crop history and diagnosis records for future monitoring and analysis.
Experimental observations show that the proposed framework provides reliable disease prediction, fast response time, and practical support for farmers. The system helps in early disease identification, reduces crop losses, improves productivity, and promotes modern smart farming practices.
Keywords: Smart Agriculture, Plant Disease Detection, Deep Learning, MobileNetV2, Chilli Leaf Classification, Precision Farming, AI Advisory System
Abstract
AI Powered Smart Banking Assistant
Dr. K Venkata Rao, Arpita Patil, Chaithanya R, Harshitha S, Jyothi P
DOI: 10.17148/IJARCCE.2026.155240
Abstract: Digital banking systems are becoming increasingly important, but many users still face difficulties while accessing banking services because of language barriers and complex interfaces. This paper proposes an AI Powered Smart Banking Assistant integrated with a dummy banking backend as BANK. The proposed system combines multilingual communication, AI chatbot support, voice assistance, and fingerprint authentication to simulate realistic banking workflows in an academic environment.
The system allows users to perform account services, ATM card services, and Aadhaar linking using voice or text interaction. A centralized MySQL database stores and manages all service requests. The proposed prototype demonstrates how AI-driven multilingual banking systems can improve accessibility and reduce dependency on manual customer support.
Keywords: Artificial Intelligence, Smart Banking, NLP, Voice Assistant, Multilingual Banking, Chatbot
The system allows users to perform account services, ATM card services, and Aadhaar linking using voice or text interaction. A centralized MySQL database stores and manages all service requests. The proposed prototype demonstrates how AI-driven multilingual banking systems can improve accessibility and reduce dependency on manual customer support.
Keywords: Artificial Intelligence, Smart Banking, NLP, Voice Assistant, Multilingual Banking, Chatbot
Abstract
Privacy-Preserving Intelligent Collaboration Platform with Searchable Encryption and Client Side AI: A Survey of Privacy-Preserving Methods for Secure Collaboration
Dr. Kothapalli Venkata Rao, Priya H M, Saniya Fatima, and Varsha R
DOI: 10.17148/IJARCCE.2026.155241
Abstract: The rapid growth of cloud-based collaboration platforms has introduced significant privacy concerns, as traditional systems expose sensitive user data to servers and third-party services during storage, indexing, and AI- based processing. Existing platforms routinely access user files for operational purposes, creating substantial risks for confidential information. This paper surveys recent privacy-preserving techniques including searchable encryption, end-to-end encryption, federated learning, and on-device AI inference, analyzing their strengths and limitations in the context of secure collaboration. Based on identified research gaps, we propose a unified Privacy-Preserving Intelligent Collaboration Platform that integrates client-side AES encryption, searchable symmetric encryption with secure tokens, RSA-based key sharing, and locally-executed AI inference to enable secure file storage, encrypted search, and intelligent features without exposing user data to servers.
Keywords: Privacy-Preserving Systems, Searchable Encryption, End-to-End Encryption, Client-Side AI, Federated Learning, Cloud Security, Secure Collaboration, AES Encryption, RSA Key Exchange, DevSecOps
Keywords: Privacy-Preserving Systems, Searchable Encryption, End-to-End Encryption, Client-Side AI, Federated Learning, Cloud Security, Secure Collaboration, AES Encryption, RSA Key Exchange, DevSecOps
Abstract
Smart Parcel Receiving System
Jayashubha J, Ashwini N R, Amrutha K, A Yashwitha, K Bindu
DOI: 10.17148/IJARCCE.2026.155242
Abstract: The increasing popularity of online shopping has created a strong need for safer and smarter parcel delivery methods. Traditional delivery systems often face issues such as parcel theft, missed deliveries, and repeated delivery attempts when customers are unavailable.
The Smart Parcel Receiving System is designed to solve these problems by using modern automation technologies. The system combines IoT, GSM communication, infrared sensors, CCTV monitoring, and LCD displays to create a secure parcel receiving environment.
When a parcel is placed inside the smart locker, the IR sensor detects the package and immediately sends a notification to the user through a mobile application or SMS. CCTV monitoring records delivery activity for additional safety and transparency. This system improves delivery efficiency, minimizes manual effort, and enhances customer convenience.
The Smart Parcel Receiving System is designed to solve these problems by using modern automation technologies. The system combines IoT, GSM communication, infrared sensors, CCTV monitoring, and LCD displays to create a secure parcel receiving environment.
When a parcel is placed inside the smart locker, the IR sensor detects the package and immediately sends a notification to the user through a mobile application or SMS. CCTV monitoring records delivery activity for additional safety and transparency. This system improves delivery efficiency, minimizes manual effort, and enhances customer convenience.
Abstract
Wildlife Poaching Risk Prediction and Detection Using Satellite AI
Mrs. Meena G, Impana P, Keerthana P, Harshitha M
DOI: 10.17148/IJARCCE.2026.155243
Abstract: Wildlife poaching is one of the major threats to biodiversity and endangered animal species across forest environments. Traditional monitoring systems based on manual patrolling and static surveillance methods are often slow and ineffective in large geographical regions. This paper proposes a Wildlife Poaching Risk Prediction and Detection System using Satellite AI. The proposed framework integrates satellite imagery analysis, environmental monitoring, machine learning algorithms, and YOLO-based suspicious activity detection for intelligent wildlife monitoring. The system supports automated risk prediction, dashboard-based monitoring, and alert generation for wildlife authorities. The framework aims to improve forest surveillance efficiency and support proactive wildlife conservation strategies.
Keywords: Wildlife Conservation, Satellite AI, Machine Learning, YOLO, Risk Prediction, Real-Time Monitoring
Keywords: Wildlife Conservation, Satellite AI, Machine Learning, YOLO, Risk Prediction, Real-Time Monitoring
Abstract
A Comparative Benchmark of Deep Learning Models and Deployment of a Web Application for Automated Early Heart Attack Risk Prediction
Ms. R.Devi, Dr. K. Padma Priya
DOI: 10.17148/IJARCCE.2026.155245
Abstract: Heart attacks remain one of the leading causes of death worldwide, highlighting the importance of early and accurate prediction. This study focuses on developing and comparing two deep learning models—CNN + MobileNet and EfficientNetV2B3—for automatic classification of ECG images and deploying the best-performing model as a real-time web application. An ECG image dataset consisting of 1,377 samples across four classes (Normal, Myocardial Infarction, Abnormal Heartbeat, and History of MI) was obtained from Kaggle. Both models were trained using TensorFlow–Keras with data augmentation and hyperparameter tuning on Google Colab (Tesla T4 GPU). Their performance was evaluated using accuracy, precision, recall, and F1-score metrics. Results showed that the CNN + MobileNet model outperformed EfficientNetV2B3, achieving 89% accuracy, 0.89 precision, 0.88 recall, and 0.89 F1- score, compared to 79% accuracy for EfficientNetV2B3. Additionally, CNN + MobileNet demonstrated smoother convergence, faster inference time (~150 ms per image), and a lightweight model size (~30 MB). Thus, CNN + MobileNet proved to be more effective for ECG classification and real-time prediction, and its deployment through a Gradio-based web interface enables accessible and rapid heart attack detection, especially in remote healthcare settings.
Keywords: Heart attack prediction, ECG classification, Deep learning, CNN + MobileNet, EfficientNetV2B3, Web deployment.
Keywords: Heart attack prediction, ECG classification, Deep learning, CNN + MobileNet, EfficientNetV2B3, Web deployment.
Abstract
Countify: A Text-to-Image Generation Model
Prof. Dr.S.S.More, Athrav Raju Ugale, Vaidehi KeshavVaze, Vaishnav Bajrang Patil, Alisha Mubarak Attar, Sanika Appasaheb Patil
DOI: 10.17148/IJARCCE.2026.155246
Abstract: This paper presents Countify, a system that integrates text-to-image generation with object detection and an iterative feedback mechanism to ensure precise object counts in generated images. The system utilizes diffusion-based image generation via ClipDrop API and object detection using YOLOv8. A validation loop continuously refines outputs until the detected object count matches the requested count.
Experimental results demonstrate that Countify significantly improves numerical accuracy in generated images, making it suitable for applications requiring precision such as dataset generation, education, and industrial automation.
Keywords: Text-to-Image Generation, Object Counting, YOLOv8, Diffusion Models, Feedback Loop, Generative AI, Computer Vision
Experimental results demonstrate that Countify significantly improves numerical accuracy in generated images, making it suitable for applications requiring precision such as dataset generation, education, and industrial automation.
Keywords: Text-to-Image Generation, Object Counting, YOLOv8, Diffusion Models, Feedback Loop, Generative AI, Computer Vision
Abstract
Real-Time Physiological Monitoring and Automated Alert System for Enhanced Soldier Safety and Survivability
Manjunath Kammar, Anushya V P, Dr. Jagadish RM, Ganta Poojith Balu, S Mounika
DOI: 10.17148/IJARCCE.2026.155247
Abstract: This research introduces the "Wounded Soldier Auto-Alert System," a prototype designed to significantly increase the survival rates of military personnel in combat zones. In high-stakes environments, the delay between an injury and medical intervention is often the difference between life and death. Our system addresses this by integrating a pulse sensor to continuously monitor critical physiological parameters—specifically vital signs, specifically pulse and internal heat. Using a Wi-Fi-enabled communication hub, this streams the readings instantly to a central processing unit. The system's primary innovation is its ability to autonomously detect anomalies; if a soldier's vital signs indicate distress, the system immediately dispatches an alert to command, complete with the soldier's pre- registered location. This moves military health monitoring from passive data collection to a proactive, life-saving tool, streamlining emergency response coordination.
Abstract
Micro-Grid Energy Management Using Inverters
Manoj Gowda B P, B J Jayadeva, Prajwal H R, Vaibhav S
DOI: 10.17148/IJARCCE.2026.155248
Abstract: This paper presents a control strategy for a solar PV–battery micro-grid using both grid-following (GFL) and grid-forming (GFM) inverters. GFL uses PLL and vector control but lacks support during grid outages. GFM, based on a virtual synchronous machine, provides voltage/frequency control, islanding, and fault ride-through capabilities. A 100 MW PV and 60 MW BESS system was simulated and validated experimentally. GFM improved system frequency by 69.3% and voltage by 70%, significantly outperforming GFL.
Keywords: grid-forming inverter; grid-following inverter; fault ride-through capability; microgrid; photovoltaic energy; power electronics
Keywords: grid-forming inverter; grid-following inverter; fault ride-through capability; microgrid; photovoltaic energy; power electronics
Abstract
AI-Based Smart Bus Live Tracking System Using IoT and Machine Learning: A Survey of Real-Time ETA Prediction and Transportation Analytics Approaches
Jayashubha J.G, Muppuri Dhatri, Raavi Pranitha, Rayapati Supriya
DOI: 10.17148/IJARCCE.2026.155250
Abstract: Public transportation systems in urban areas suffer from persistent inefficiencies rooted in static schedule management, absent live vehicle visibility, and reactive fleet operations. Passengers endure uncertainty regarding bus arrival times, leading to excessive stop-level waiting and reduced confidence in public transit. This paper surveys existing research on IoT-enabled bus tracking, AI-based estimated time of arrival (ETA) prediction, GPS-based fleet monitoring, and machine learning applied to transportation analytics. Six representative studies from 2022–2026 are analyzed and compared across methodology, hardware configuration, AI algorithm, cloud platform, and key limitations. Based on this survey, we propose a comprehensive AI-Based Smart Bus Live Tracking System integrating ESP32 microcontrollers, GPS modules, cloud-synchronized databases, machine learning-based ETA prediction, and a cross-platform Flutter passenger application. The proposed system addresses critical gaps in existing approaches by combining real-time GPS tracking, multi-parameter AI prediction, and a centralized analytics dashboard within a single deployable platform.
Keywords: IoT, GPS, ESP32, ETA Prediction, Machine Learning, Smart Transportation, Real-Time Tracking, Cloud Computing, Flutter, Transportation Analytics, LSTM, Random Forest, Firebase, Smart City
Keywords: IoT, GPS, ESP32, ETA Prediction, Machine Learning, Smart Transportation, Real-Time Tracking, Cloud Computing, Flutter, Transportation Analytics, LSTM, Random Forest, Firebase, Smart City
Abstract
AI-Based Intelligent Document a Verification System Using OCR and Machine Learning
Vrushali Gavali, Rutuja Shelar, Pranita Sandim
DOI: 10.17148/IJARCCE.2026.155251
Abstract: In the modern digital era, document verification has become an essential requirement across various sectors such as education, banking, government, and corporate industries. Traditional verification methods rely heavily on manual processes, which are time-consuming, error-prone, and inefficient in handling large volumes of data. This paper presents a Document Verification System that automates the process of validating documents using advanced technologies. The system aims to reduce human intervention, improve verification accuracy, and provide faster results. The proposed system allows users to upload documents through a web-based interface developed using ReactJS, with state management implemented using Redux/Context API and UI styling using TailwindCSS/Bootstrap. The backend is developed using Python, which processes the uploaded documents and verifies them against predefined data stored in MongoDB/MySQL databases. The system incorporates TensorFlow-based machine learning models and Optical Character Recognition (OCR) techniques to extract and analyze document content for authenticity.
Keywords: Document Verification, OCR, Machine Learning, Deep Learning, Fraud Detection, TensorFlow, ReactJS, Artificial Intelligence, Authentication System.
Keywords: Document Verification, OCR, Machine Learning, Deep Learning, Fraud Detection, TensorFlow, ReactJS, Artificial Intelligence, Authentication System.
Abstract
Artificial Intelligence Techniques in Clinical Decision Support Systems in Radiology Using X-ray and MRI
Shubhmehar Singh Dhillon*, Sandeep Kaur, Satveer Kour
DOI: 10.17148/IJARCCE.2026.155252
Abstract: Artificial Intelligence (AI) is changing the Clinical Decision Support Systems (CDSS) in radiology, especially in X-ray and Magnetic Resonance Imaging (MRI). This study involves a qualitative systematic synthesis of 70 relevant publications with PRISMA approach to assess AI approaches, clinical applications and real-world deployment in radiological CDSS. The findings show that deep learning models like CNNs, 3D CNNs, U-Net, transformer-based models dominate in today's applications. The advantage of AI-CDSS based on X-rays are for screening and classification, whilst AI-CDSS based on MRI are better for volumetric analysis, tumour segmentation, and data monitoring. Aidoc, Viz.ai, Qure.ai, Arterys and DeepMind Health demonstrate the therapeutic use of AI in triage, anomaly detection and workflow optimisation. However, challenges remain concerning interpretability, data heterogeneity, generalisability and clinical integration. AI is rapidly evolving as a vital companion in radiological CDSS, improving accuracy and efficiency.
Keywords: Artificial Intelligence, Clinical Decision Support Systems, Radiology, X-ray, MRI, Deep Learning, Explainable AI
Keywords: Artificial Intelligence, Clinical Decision Support Systems, Radiology, X-ray, MRI, Deep Learning, Explainable AI
Abstract
AI Powered Mental Health Companion
Hasini P, Jestina M, D.Sasmitha M.E
DOI: 10.17148/IJARCCE.2026.155253
Abstract: Mental health issues such as stress, anxiety, and depression are increasing rapidly among students and working professionals. Many individuals hesitate to seek professional help due to fear, lack of awareness, and high consultation costs. This project proposes an AI Powered Mental Health Companion that provides emotional support using Artificial Intelligence and Natural Language Processing techniques. The system analyzes user emotions, provides supportive responses, and recommends wellness activities. The proposed system helps users manage emotional stress and improve mental well-being through intelligent interaction.
Keywords: Artificial Intelligence, Mental Health, Emotion Detection, NLP, Chatbot, Machine Learning, Stress Management
Keywords: Artificial Intelligence, Mental Health, Emotion Detection, NLP, Chatbot, Machine Learning, Stress Management
Abstract
AI-Based Virtual Interview Assistant: An Intelligent NLP-Driven System for Automated Interview Evaluation and Performance Analysis
Anitha, Ankith Kumar Verma, Archana Kulkarni, B S Shree Roopa, Dr. Phanindra Reddy K
DOI: 10.17148/IJARCCE.2026.155254
Abstract: The rapid growth of competitive recruitment processes has increased the demand for intelligent interview prepa-ration platforms capable of providing personalized guidance and automated performance evaluation. Traditional mock interview systems depend heavily on trainers and manual evaluators, making the process time-consuming, inconsistent, and inaccessible to many students and job seekers. This paper presents an AI-Based Virtual Interview Assistant that automates interview simulation, response analysis, and performance assessment using Natural Language Processing (NLP) and speech processing techniques. The proposed system allows candidates to participate in domain- specific mock interviews using text or voice responses. Candidate answers are evaluated using semantic similarity analysis, TF-IDF vectorization, cosine similarity scoring, keyword matching, sentiment analysis, and communication assessment techniques. Speech responses are processed using Speech-to-Text conversion and analyzed for fluency, pauses, clarity, and confidence level. Based on the evaluation results, the system generates auto-mated feedback reports, performance scores, and improvement recommendations. The system was implemented using Python, Flask, PostgreSQL, spaCy, NLTK, HTML, CSS, JavaScript, and Tailwind CSS. Experimental evaluation demonstrates that the proposed system provides scalable, consistent, and intelligent interview preparation support while significantly reducing de-pendency on human evaluators.
Keywords: Artificial Intelligence, Natural Language Processing, Interview Assistant, Mock Interview System, Speech Analysis, Semantic Evaluation, Automated Feedback.
Keywords: Artificial Intelligence, Natural Language Processing, Interview Assistant, Mock Interview System, Speech Analysis, Semantic Evaluation, Automated Feedback.
Abstract
A Comparative Analysis of Lightweight Deep Learning Models for Real-Time Apple Orchard Monitoring
Mansi Sharma
DOI: 10.17148/IJARCCE.2026.155255
Abstract: Apple cultivation represents a significant global investment; profit losses result when growers are unable to detect leaf disease at early stages. Deep learning has accelerated disease diagnosis, although a deployment gap for mobile devices remains. The majority of models are not efficient enough to run on the standard smartphones that farmers actually own. For this reason, this research evaluates the practical suitability of three deep learning models: EfficientNet-B0, MobileNetV2 and VGG16. To determine the actual trade-offs between the accuracy and power drain of a model, this research trained each model on a dataset containing apple scab, black rot, and cedar apple rust. The highest validation accuracy achieved by EfficientNet-B0 was 97.8% and F1-score of 0.977. MobileNetV2, however, was more feasible for edge deployment owing to its far fewer parameters. This study provides a scalable, real-time apple orchard monitoring solution by evaluating the on-device performance of these models.
Keywords: Apple leaf disease detection, Deep learning, EfficientNet-B0, MobileNetV2, Lightweight model, Apple scab, Black rot, Precision agriculture.
Keywords: Apple leaf disease detection, Deep learning, EfficientNet-B0, MobileNetV2, Lightweight model, Apple scab, Black rot, Precision agriculture.
Abstract
Green Elixir Vision: An Intelligent Ayurvedic Herbal Recommendation Platform for Personalized Natural Healthcare
H. Srinivas, J. Vinay, K. Basavana Gouda, K. G Ravi Teja Gowda, Dr. C. N. Shariff
DOI: 10.17148/IJARCCE.2026.155256
Abstract: The increasing dependence on synthetic medications for minor health conditions has created significant concerns regarding long-term side effects, reduced immunity, and antibiotic resistance. Traditional medical systems such as Ayurveda provide effective and natural alternatives; however, access to authenticated herbal knowledge remains fragmented across books, practitioners, and unstructured online resources.
This paper presents Green Elixir Vision, an intelligent web- based herbal healthcare platform that combines Ayurvedic medicinal knowledge with Artificial Intelligence techniques to provide personalized herbal recommendations. The proposed system integrates Natural Language Processing, symptom interpretation, rule-based recommendation logic, and a structured medicinal plant database to assist users in identifying suitable herbal remedies for common health issues.
Experimental evaluation demonstrates reliable recommendation accuracy, efficient data retrieval, and improved user accessibility.
Keywords: Ayurveda, Artificial Intelligence, Herbal Recommendation, Natural Language Processing, Personalized Health- care, Medicinal Plants
This paper presents Green Elixir Vision, an intelligent web- based herbal healthcare platform that combines Ayurvedic medicinal knowledge with Artificial Intelligence techniques to provide personalized herbal recommendations. The proposed system integrates Natural Language Processing, symptom interpretation, rule-based recommendation logic, and a structured medicinal plant database to assist users in identifying suitable herbal remedies for common health issues.
Experimental evaluation demonstrates reliable recommendation accuracy, efficient data retrieval, and improved user accessibility.
Keywords: Ayurveda, Artificial Intelligence, Herbal Recommendation, Natural Language Processing, Personalized Health- care, Medicinal Plants
Abstract
Web Security Issues Analysis and Practical Ways to Prevent Them
Praveen kumar, Manjeet kaur
DOI: 10.17148/IJARCCE.2026.155257
Abstract: Web applications have become an important part of our daily life and are widely used in communication, education, banking, shopping and business services. The use of web applications is increasing and also the security related problems are increasing rapidly. A lot of applications are vulnerable to SQL Injection, Cross-Site Scripting (XSS), weak authentication mechanism and insecure configuration settings. Such vulnerabilities can lead to unauthorised access, theft of sensitive information, or disruption of system operations, with serious financial and privacy-related consequences.
The main objective of this research is to study the security challenges present in modern web applications and to understand the impact of these vulnerabilities on data protection and system reliability. For this purpose, we shall utilise a real-world dataset collected from Kaggle for analysis. The aim of the study is to determine the frequency and distribution of different vulnerabilities and to look for the reasons of their appearance. It tries to find out which security problems are most frequent in development and maintenance of web applications [1].
The analysis of the dataset will allow the research to identify the areas most in need of security improvements and the factors that contribute to these weaknesses. The results will be used to recommend effective preventive measures and security practices for the reduction of risks and the improvement of the overall security of web applications [1].
Keywords: Web Application Security, SQL Injection, Cross-Site Scripting (XSS), Authentication, Cybersecurity, Vulnerability Analysis, Data Protection, Secure Web Development, Kaggle Dataset, Cyber Security
The main objective of this research is to study the security challenges present in modern web applications and to understand the impact of these vulnerabilities on data protection and system reliability. For this purpose, we shall utilise a real-world dataset collected from Kaggle for analysis. The aim of the study is to determine the frequency and distribution of different vulnerabilities and to look for the reasons of their appearance. It tries to find out which security problems are most frequent in development and maintenance of web applications [1].
The analysis of the dataset will allow the research to identify the areas most in need of security improvements and the factors that contribute to these weaknesses. The results will be used to recommend effective preventive measures and security practices for the reduction of risks and the improvement of the overall security of web applications [1].
Keywords: Web Application Security, SQL Injection, Cross-Site Scripting (XSS), Authentication, Cybersecurity, Vulnerability Analysis, Data Protection, Secure Web Development, Kaggle Dataset, Cyber Security
Abstract
A Research on Machine Learning Driven Gamification Model for Personalized Education
Sadiya Ali, Dr. G. R. Bamnote, Dr. G. J. Sawale
DOI: 10.17148/IJARCCE.2026.155258
Abstract: Personalized learning has become an important approach in modern education as traditional learning systems often fail to address the individual needs, abilities, and learning pace of students. At the same time, maintaining student motivation and engagement in online learning environments remains a major challenge. The proposed system integrates personalized learning recommendations, adaptive difficulty adjustment, reward optimization, and engagement tracking into a unified web-based learning platform. Machine learning models analyse student performance, quiz history, learning behaviour, and engagement data to recommend suitable quizzes and courses according to the learner’s skill level. The system categorizes recommendations into easier, same-level, and harder learning paths to support adaptive learning experiences.
To increase motivation and participation, the platform incorporates gamification features such as points, badges, levels, leaderboards, and achievement tracking. Student activities including logins, lesson views, and quiz attempts are continuously monitored to evaluate engagement and provide personalized feedback. In addition, a teacher analytics dashboard is implemented to help educators monitor student progress, identify at-risk learners, and analyse academic performance using predictive insights generated through machine learning.
Experimental evaluation and literature-supported analysis indicate that the integration of machine learning and gamification improves learner engagement, supports self-directed learning, enhances personalization, and contributes to better academic outcomes.
Keywords: Machine Learning, Personalized Education, Gamification, Gamified Learning, Gamification Model
To increase motivation and participation, the platform incorporates gamification features such as points, badges, levels, leaderboards, and achievement tracking. Student activities including logins, lesson views, and quiz attempts are continuously monitored to evaluate engagement and provide personalized feedback. In addition, a teacher analytics dashboard is implemented to help educators monitor student progress, identify at-risk learners, and analyse academic performance using predictive insights generated through machine learning.
Experimental evaluation and literature-supported analysis indicate that the integration of machine learning and gamification improves learner engagement, supports self-directed learning, enhances personalization, and contributes to better academic outcomes.
Keywords: Machine Learning, Personalized Education, Gamification, Gamified Learning, Gamification Model
Abstract
QR CODE BASED CAFETERIA FOOD ORDERING AND PAYMENT SYSTEM
Dr. V. U. Bansude, Pathan Surayya, Mane Gauri, Patale Mayuri
DOI: 10.17148/IJARCCE.2026.155259
Abstract: Traditional methods of placing food orders and settling bills in restaurants often lead to confusion, errors, and unnecessary delays. To address these challenges, a QR code-based food ordering system has been proposed to streamline the ordering process. Customers can scan a QR code available at their table using a smartphone, which instantly provides access to the restaurant’s menu. They can then select and customize their orders without waiting for a waiter. This system also allows secure and convenient bill settlement directly through the customer’s device, enhancing overall efficiency. The traditional approach—where waiters manually take orders using pen-and- paper or printed menus—can lead to miscommunication between staff and customers, and often requires customers to wait for service. By leveraging QR code technology, these issues can be mitigated, providing a faster, more accurate, and user-friendly ordering experience.
Moreover, the system benefits restaurant staff by enabling easy management of menus and real- time tracking of orders. Since smartphones are widely used today, implementing such a system ensures accessibility for nearly all customers. Overall, the QR code-based food ordering system offers a practical and modern solution to improve both customer satisfaction and operational efficiency in restaurants.
Keywords: QR Code Ordering System Digital Menu ESP32 Microcontroller Contactless Food Ordering Smart Cafeteria IoT-based Food Service Online Ordering System Real-time Order Management Automated Billing Customer Experience in Cafeterias Embedded Systems in Food Service Wireless Communication in Cafeterias.
Moreover, the system benefits restaurant staff by enabling easy management of menus and real- time tracking of orders. Since smartphones are widely used today, implementing such a system ensures accessibility for nearly all customers. Overall, the QR code-based food ordering system offers a practical and modern solution to improve both customer satisfaction and operational efficiency in restaurants.
Keywords: QR Code Ordering System Digital Menu ESP32 Microcontroller Contactless Food Ordering Smart Cafeteria IoT-based Food Service Online Ordering System Real-time Order Management Automated Billing Customer Experience in Cafeterias Embedded Systems in Food Service Wireless Communication in Cafeterias.
Abstract
A Machine Learning-based Crop Yield Forecasting and Recommendation System
C M Abhishek, Channaveeresha Meti, Revanasiddappa, Mallikarjun Kallappa Poojari, Dr. Jagadish R M, Manjunath Kammar
DOI: 10.17148/IJARCCE.2026.155260
Abstract: Reliable agricultural forecasting is essential for effective crop planning, yield estimation, and climate-aware decision making, particularly in regions where farming activity is closely tied to seasonal rainfall. This paper presents an integrated machine learning-based agricultural prediction system that addresses three core tasks: crop selection, crop yield forecasting, and rainfall estimation using district-level datasets from Karnataka, India.
The proposed system employs optimized tree-based ensemble models, including Random Forest, XGBoost, LightGBM, and CatBoost, which are well suited for structured agricultural data commonly available in developing regions. Instead of relying on deep sequence models that require long temporal weather records and genotype-related inputs, the system operates on seasonal and regional attributes, enabling efficient training and deployment on modest hardware. The trained models are integrated into a web-based portal developed using PHP, Python, and MySQL, allowing farmers to access predictions through simple and intuitive interfaces. Comparative evaluation against a recent LSTM-based yield prediction framework shows that the proposed system achieves competitive or improved accuracy while remaining significantly more practical for real-world deployment.
Keywords: Agriculture, Machine Learning, Crop Prediction, Yield Forecasting, Rainfall Prediction, Web Application, Decision Support System.
The proposed system employs optimized tree-based ensemble models, including Random Forest, XGBoost, LightGBM, and CatBoost, which are well suited for structured agricultural data commonly available in developing regions. Instead of relying on deep sequence models that require long temporal weather records and genotype-related inputs, the system operates on seasonal and regional attributes, enabling efficient training and deployment on modest hardware. The trained models are integrated into a web-based portal developed using PHP, Python, and MySQL, allowing farmers to access predictions through simple and intuitive interfaces. Comparative evaluation against a recent LSTM-based yield prediction framework shows that the proposed system achieves competitive or improved accuracy while remaining significantly more practical for real-world deployment.
Keywords: Agriculture, Machine Learning, Crop Prediction, Yield Forecasting, Rainfall Prediction, Web Application, Decision Support System.
Abstract
A Review Study of Quantum Computing
Gurpreet Kaur, Satveer Kour, Sandeep Kaur
DOI: 10.17148/IJARCCE.2026.155261
Abstract: A new paradigm of computation has appeared recently that is known as quantum computing and which is able to solve difficult problems that cannot be solved using classical computing. It provides a comprehensive overview of the basic concepts, computational models, architectures, tools, applications and implications of quantum computing. The paper is based on the latest research, and it's used to evaluate the state of quantum computing, from theory to practical, application-oriented implementations. Fundamental concepts such as qubits, superposition, entanglement and hybrid quantum-classical computation are explained, along with major models and frameworks of quantum computing, including Cirq, TensorFlow Quantum and ProjectQ. Furthermore, various significant application areas, such as cryptography, optimization, AI, drug discovery, and industrial computing, are explored, highlighting the applications' potential and their technological development. Although great strides have been made, scalability, noise, error correction, interoperability and hardware challenges are major problems in the NISQ era. It also examines new trends, future research directions and the societal, security and ethical considerations of the advancement of quantum technology. But the bigger picture indicates that quantum computing is progressing from theory to reality, and there are many technical and institutional challenges to be resolved along the way. Interdisciplinary studies and responsible innovation will be key to unlocking the potential of quantum computing as the next generation of computational technology.
Keywords: Quantum Computing, Quantum Algorithms, Qubits, Quantum Machine Learning, Quantum Optimization, Noisy Intermediate-Scale Quantum (NISQ).
Keywords: Quantum Computing, Quantum Algorithms, Qubits, Quantum Machine Learning, Quantum Optimization, Noisy Intermediate-Scale Quantum (NISQ).
Abstract
From Intuition to Agriculture 4.0: A Comprehensive Review of Internet of Things (IoT) and Machine Learning in Precision Farming
Maryen, Satinder Kaur
DOI: 10.17148/IJARCCE.2026.155262
Abstract: The global agricultural sector is currently undergoing a massive paradigm shift, transitioning from traditional, intuition-based farming practices to highly optimized, data-driven methodologies. Driven by rapid population growth and unpredictable climate volatility, the emergence of "Agriculture 4.0" relies heavily on the integration of the Internet of Things (IoT) and Machine Learning (ML). This paper provides a comprehensive review of the modern precision agriculture ecosystem. It explores the architectural flow of wireless sensor networks, the role of multi-source data acquisition (including UAVs and soil nodes), and the foundational transition toward algorithmic decision-making. Ultimately, this review establishes that while IoT hardware provides the necessary biological data, the future of smart farming depends entirely on the development of advanced computational software capable of translating environmental metrics into actionable crop recommendations.
Abstract
Beyond Standalone Classifiers: A Critical Review of Multi-Paradigm and Ensemble Machine Learning Architectures in Crop Recommendation Systems
Maryen, Satinder Kaur
DOI: 10.17148/IJARCCE.2026.155263
Abstract: As precision agriculture transitions toward highly diversified, data-driven farming environments, the limitations of traditional predictive modeling are becoming increasingly apparent. Historically, crop recommendation engines have relied on standalone machine learning classifiers, such as Logistic Regression, Support Vector Machines (SVM), and K-Nearest Neighbors (KNN). While effective in narrow, localized datasets, these monolithic algorithms consistently fail to capture the complex, non-linear biological synergies required for high-diversity crop matrices. This paper provides a critical review of the evolution of agricultural machine learning. It explores the structural limitations of early linear models, analyzes the shift toward tree-based boosting algorithms (like XGBoost), and ultimately argues that the future of precision agronomy relies entirely on multi-paradigm, hybrid ensemble architectures capable of bypassing the traditional bias-variance tradeoff.
Abstract
ROLE AND CHALLENGES OF HARITHA KARMA SENA: A STUDY IN ERNAKULAM DISTRICT
Dr. Francis M C, Anosh Ignatious, Diya Damian and Dr. Somasekharan T M
DOI: 10.17148/IJARCCE.2026.155264
Abstract: Waste management has become a major environmental challenge due to rapid urbanisation, population growth, and increasing waste generation. In Kerala, Haritha Karma Sena plays an important role in decentralised waste management through activities such as waste collection, segregation, recycling, and awareness creation. This study examines the role and challenges of Haritha Karma Sena in Ernakulam district with special focus on operational difficulties, waste-to-wealth initiatives, technological adoption, and the promotion of sustainable alternatives to plastic. The study is based on both primary and secondary data collected through structured questionnaires, research articles, journals, and government reports. The findings reveal that Haritha Karma Sena contributes significantly to environmental protection and sustainable waste management, but faces challenges such as low income, inadequate safety measures, heavy workload, limited infrastructure, and insufficient public cooperation. The study also highlights the importance of community participation, technological support, and eco-friendly practices in improving the effectiveness of decentralised waste management systems. The research provides useful suggestions for strengthening Haritha Karma Sena activities and promoting sustainable environmental practices in Ernakulam district.
Keywords: Haritha Karma Sena, Waste Management, Decentralised Waste Management, Waste-to-Wealth, Sustainable Development
Keywords: Haritha Karma Sena, Waste Management, Decentralised Waste Management, Waste-to-Wealth, Sustainable Development
Abstract
INTELLIGENT SCREAM DETECTION SYSTEM FOR CRIME ALERTS
Dr.Hemavati C Purad, Putti Monashree, Pawan Kumar N, Sai Ganesh M
DOI: 10.17148/IJARCCE.2026.155265
Abstract: In recent times, personal safety has become a serious concern, especially in situations where individuals are unable to call for help. This research focuses on developing an intelligent system that can automatically detect human screams and send alerts during emergencies. The proposed system uses audio processing and deep learning techniques to identify distress signals and notify emergency contacts without requiring manual action. By reducing response time and ensuring faster communication, the system aims to provide a reliable and practical safety solution. This approach can be useful in real-life scenarios such as women’s safety, elderly care, and public surveillance.
Keywords: Intelligent Scream Detection, Audio Signal Processing, Deep Neural Networks (DNN), MFCC, Emergency Alert Systems, Real-Time Monitoring, Public Safety.
Keywords: Intelligent Scream Detection, Audio Signal Processing, Deep Neural Networks (DNN), MFCC, Emergency Alert Systems, Real-Time Monitoring, Public Safety.
Abstract
Correlation and Regression: A Comprehensive Review of Statistical Relationships and Predictive Modeling
Anagha Bade
DOI: 10.17148/IJARCCE.2026.155266
Abstract: Correlation and regression are among the most important statistical techniques used in mathematics, economics, business, and data science for analyzing relationships between variables. Correlation measures the strength and direction of association between variables, while regression establishes predictive relationships and quantifies the effect of independent variables on dependent variables. This review paper discusses the theoretical foundations, types, methods, assumptions, applications, similarities, and differences between correlation and regression analysis. The study further explains how these techniques are applied in forecasting, decision-making, econometrics, and financial analysis. The paper concludes that correlation identifies associations, whereas regression provides predictive and explanatory insights for practical applications.
Keywords: Correlation, Regression, Pearson Correlation, Linear Regression, Statistical Analysis, Predictive Modeling, Econometrics.
Keywords: Correlation, Regression, Pearson Correlation, Linear Regression, Statistical Analysis, Predictive Modeling, Econometrics.
Abstract
AI‑ML Powered Robot for OCR‑Driven Payment Reconciliation using UiPath
Ashutosh Mankar, Dr G.R Bamnote, Prof. S.P Akarte
DOI: 10.17148/IJARCCE.2026.155267
Abstract: This research paper presented an AI ML powered robot for OCR driven payment reconciliation using UiPath. The study used an open source Accounts Receivable dataset exported from an ERP system, where customer balances were recalculated from invoices, payments, credits, and adjustments to identify integration errors. The project used three scripts for synthetic data generation, rule based reconciliation, and anomaly detection through Linear Regression. The UiPath based automation was developed with Outlook, Excel, System, and DataTable activities, along with reusable workflows for email reading, OCR extraction, file validation, data cleaning, matching, exception handling, and output generation. Extracted information from invoices and receipts was converted to structured Excel output to be reviewed and reported on. The results showed significant changes in operations including 78.57% less time spent processing invoices, 65% less time spent preparing orders, 66.67% less time spent reconciling inventory payments, 90% less time spent making mistakes when matching invoices and 20 to 25% less money spent on operations.
Keywords: AI ML, OCR, UiPath, Payment Reconciliation, Robotic Process Automation, Invoice Processing.
Keywords: AI ML, OCR, UiPath, Payment Reconciliation, Robotic Process Automation, Invoice Processing.
Abstract
Design And Implementation of An Embedded CNN based Weed Detection and Mechanical Weed Elimination Rover
Sai Badgujar, Raj Patil, Gaurav Shinde, Parth Borgude, Rupali Shirsath, Prof. Mayur Kumbharde
DOI: 10.17148/IJARCCE.2026.155268
Abstract: Weed invasion remains a significant challenge in modern agriculture, contributing to reduced crop productivity and increased operational costs. Conventional weed control practices, including manual weeding and chemical herbicide application, are labour-intensive and raise environmental and health concerns. This study presents the development, implementation, and experimental assessment of an embedded deep learning system based on computer vision for identifying weeds and removing them mechanically using a mobile robotic platform.
The proposed system employs a Convolutional Neural Network (CNN) deployed implemented on a compact embedded device like the Raspberry Pi to perform on-device weed detection from camera-acquired field images. Detected weeds are localized and processed in real time to guide a servo-driven mechanical plucking mechanism, enabling physical weed removal without the use of chemical herbicides. The system integrates visual perception, embedded processing, rover mobility, and mechanical actuation into a compact and low-cost platform intended for small- and medium-scale agricultural settings.
Experimental evaluation was conducted under controlled and limited real-field conditions to assess detection performance and system feasibility. The outcome indicate that the developed system can successfully identify weeds and perform targeted mechanical removal while operating within the computational constraints of embedded hardware. These findings demonstrate the capability of embedded deep learning robotic systems to support precise and chemical-free weed control solutions.
Keywords: Weed Detection, Convolutional Neural Networks, Precision Agriculture, Agricultural Robotics, Autonomous Rover, Edge AI, Mechanical Weed Removal.
The proposed system employs a Convolutional Neural Network (CNN) deployed implemented on a compact embedded device like the Raspberry Pi to perform on-device weed detection from camera-acquired field images. Detected weeds are localized and processed in real time to guide a servo-driven mechanical plucking mechanism, enabling physical weed removal without the use of chemical herbicides. The system integrates visual perception, embedded processing, rover mobility, and mechanical actuation into a compact and low-cost platform intended for small- and medium-scale agricultural settings.
Experimental evaluation was conducted under controlled and limited real-field conditions to assess detection performance and system feasibility. The outcome indicate that the developed system can successfully identify weeds and perform targeted mechanical removal while operating within the computational constraints of embedded hardware. These findings demonstrate the capability of embedded deep learning robotic systems to support precise and chemical-free weed control solutions.
Keywords: Weed Detection, Convolutional Neural Networks, Precision Agriculture, Agricultural Robotics, Autonomous Rover, Edge AI, Mechanical Weed Removal.
Abstract
AI-Based Sign Language Recognition System
Dr. K. Prem Kumar, S. Pranavi, T. Manojghna, P. Manaswini, Y. Bhavitha
DOI: 10.17148/IJARCCE.2026.155269
Abstract: The communication gap between the hearing-impaired community and the hearing population remains a significant challenge, restricting access to seamless integration in modern educational, professional, and social environments. Traditional methods, such as the employment of human sign language interpreters, are often financially prohibitive, geographically unavailable, or limited by real-time processing constraints. This project proposes an innovative AI-Based Sign Language Recognition System that leverages the synergistic capabilities of Deep Learning, Computer Vision, and Convolutional Neural Networks (CNN) to bridge this accessibility gap. Our methodology integrates high-fidelity hand-tracking through the MediaPipe framework, coupled with a custom-trained CNN model specifically optimized for classifying complex sign language alphabets in real- time. The system pipeline includes a robust image processing backbone, a scalable gesture prediction engine, and an intuitive web-based interface that delivers instant text and speech feedback to the user. Extensive performance evaluations and rigorous testing demonstrate that the proposed system achieves exceptional classification accuracy and maintains low-latency inference, even under varying ambient lighting conditions. By providing a scalable, portable, and user-friendly communication tool, this research serves as a viable, modern alternative to expensive sensor-glove hardware, establishing a new foundation for inclusive assistive technologies in digital education.
Keywords: Artificial Intelligence, Computer Vision, Deep Learning, CNN, MediaPipe, OpenCV, Gesture Detection, Real-time Sign Language Recognition, Assistive Technology.
Keywords: Artificial Intelligence, Computer Vision, Deep Learning, CNN, MediaPipe, OpenCV, Gesture Detection, Real-time Sign Language Recognition, Assistive Technology.
Abstract
Predictive Analysis of Health and Lifestyle Patterns using CRM and Machine Learning
Prathamesh Chambole, Dr S.R Gupta, Dr R.A Kale
DOI: 10.17148/IJARCCE.2026.155270
Abstract: This study shows how CRM and machine learning can be used to predict health and living trends in order to help with preventive healthcare. The study uses an open source dataset that has factors about demographics, physical exercise, food, sleep, mental health, and medical background. Data preparation was done to make the data better by dealing with missing values, getting rid of duplicate and incomplete records, lowering the number of errors, and getting the dataset ready for reliable model training. It was possible to turn clinical, physiological, and behavioral traits into useful data for machine learning models by using feature engineering and feature extraction. For figuring out health risks and predicting them, Random Forest, Decision Tree, Logistic Regression, XGBoost, and voting-based classification methods were looked at. With AUC values of 0.68 and 0.64, respectively, the ROC results showed that Random Forest did a little better than Decision Tree. The overall risk classification showed that 43.7 percent of people were low risk, 35.4% were intermediate risk, and 20.9 percent were high risk.
Keywords: Health and lifestyle patterns; Machine learning; Preventive healthcare; Random Forest; Decision Tree; Risk stratification.
Keywords: Health and lifestyle patterns; Machine learning; Preventive healthcare; Random Forest; Decision Tree; Risk stratification.
Abstract
A Lightweight HSV Histogram-Based Algorithm for Real-Time Face Recognition on Edge Devices
Prof. Roshni Gawande, Dr. S. B. Patil, Prof. Sneha Dhere
DOI: 10.17148/IJARCCE.2026.155271
Abstract: Real-time face recognition systems are increasingly deployed in mobile and edge environments, yet most deep learning approaches demand GPU acceleration and large memory footprints. This paper presents a lightweight algorithm combining Haar Cascade detection, HSV histogram feature extraction, and cosine similarity classification. Implemented via a Flutter mobile client and Flask REST API backend, the system achieves an average end-to-end latency of 82.5 ms and a False Acceptance Rate (FAR) of 0.25% without GPU support. Experimental evaluation demonstrates that algorithmic efficiency and minimal infrastructure overhead can outweigh marginal accuracy gains of deep learning models in controlled access-control scenarios.
Abstract
Kanwariya Sewa: A Smart Logistics and Wellness Platform for Pilgrimage Support Using Real-Time Tracking and Emergency Response System
Subhash, Swati Sinha, Shrey Agnihotri, Dr. Deepak Kumar Gupta
DOI: 10.17148/IJARCCE.2026.155272
Abstract: Large-scale religious pilgrimages such as the Kanwar Yatra involve millions of pilgrims traveling across long routes, which creates challenges in crowd management, emergency response, health support, and facility coordination. Traditional management systems mainly depend on manual monitoring and lack real-time digital support. This paper presents “Kanwariya Sewa”, a smart logistics and wellness platform designed to improve pilgrim safety and coordination using modern web technologies. The proposed system provides features such as pilgrim registration, real-time location tracking, SOS emergency alerts, facility mapping, and centralized administration. The platform is developed using React.js for the frontend, Node.js and Express.js for the backend, and MongoDB as the database. Google Maps API is integrated for route and location services.
The proposed system improves emergency response time, enhances coordination between pilgrims and volunteers, and provides a scalable digital solution for pilgrimage management aligned with Digital India initiatives.
Keywords: Pilgrimage Management, Emergency Response System, Real-Time Tracking, Smart Logistics, Kanwar Yatra, Web Application
The proposed system improves emergency response time, enhances coordination between pilgrims and volunteers, and provides a scalable digital solution for pilgrimage management aligned with Digital India initiatives.
Keywords: Pilgrimage Management, Emergency Response System, Real-Time Tracking, Smart Logistics, Kanwar Yatra, Web Application
Abstract
Comparative Performance Analysis of Quantum Error Correction Codes under Simulated Depolarizing Noise using Qiskit
Gurpreet Kaur, Satveer Kour, Sandeep Kaur, Satinder Kaur
DOI: 10.17148/IJARCCE.2026.155273
Abstract: Fault-tolerant quantum computation requires quantum error correction (QEC), and while the performance of various QEC codes has been well characterized under idealized noise conditions, it has not been sufficiently characterized by simulation under realistic noise conditions. In this paper, the performance of three basic QEC codes (3-qubit Bit-Flip, 3-qubit Phase-Flip and Shor's 9-qubit [[9,1,3]]) under simulated depolarizing noise, by using Qiskit Aer, is analyzed systematically in three stages. Stage 1 is a validation of all three implementations under ideal circumstances. Stage 2 tests for 6 different error rates of 0% to 5%, and runs 2048 shots per error rate. Stage 3 brings the analysis from stage 2 to 0% to 10%, but with an uncorrected baseline for comparison. Results verify that Shor's > Bit-Flip > Phase-Flip is consistent over practically relevant error rates. At 1% noise, within current IBM Quantum hardware range, all three codes achieve a logical fidelity of over 98.7%. An error rate of 7–8% is used to identify a cross-over threshold above which the simpler Bit-Flip code can match Shor's code in fidelity with only one-third the number of qubits.
Keywords: Quantum Error Correction, Depolarizing Noise, Bit-Flip Code, Phase-Flip Code, Shor's Code, Logical Fidelity, Qiskit, NISQ, Fault-Tolerant Quantum Computing, Error Correction Threshold.
Keywords: Quantum Error Correction, Depolarizing Noise, Bit-Flip Code, Phase-Flip Code, Shor's Code, Logical Fidelity, Qiskit, NISQ, Fault-Tolerant Quantum Computing, Error Correction Threshold.
Abstract
Zero Trust Architecture in Cloud Security: Principles, Implementation, and Challenges
Varun Kumar, Sandarsh Gowda M M
DOI: 10.17148/IJARCCE.2026.155274
Abstract: The rapid proliferation of cloud computing has fundamentally transformed the way organizations design, deploy, and manage their IT infrastructure. Traditional perimeter-based security models, which rely on the concept of a trusted internal network, are no longer adequate in an era of distributed workloads, remote access, and advanced persistent threats. Zero Trust Architecture (ZTA) has emerged as a transformative security paradigm that operates on the principle of "never trust, always verify," treating every user, device, and network flow as potentially hostile regardless of its origin. This paper presents a comprehensive examination of Zero Trust Architecture in the context of cloud security, covering its foundational principles, practical implementation frameworks, and the technical and organizational challenges encountered during adoption. The study reviews established models such as NIST SP 800-207 and the Forrester Zero Trust eXtended (ZTX) framework, and analyzes core ZTA components including micro-segmentation, identity and access management (IAM), continuous monitoring, multi-factor authentication (MFA), and least-privilege access control. Furthermore, this research explores real-world deployment case studies across multi-cloud and hybrid environments and identifies key barriers such as legacy system integration, complexity of policy management, and latency concerns. The findings indicate that while ZTA implementation demands significant organizational and technical investment, it substantially reduces the attack surface and enhances resilience against modern cloud-based threats. The paper concludes with recommendations for a phased ZTA adoption roadmap suited to organizations at varying levels of cloud maturity. Cloud computing has revolutionized modern enterprise infrastructure by providing scalable, flexible, and cost-efficient computing resources across distributed environments. However, the increasing adoption of cloud platforms has also introduced significant cybersecurity challenges due to remote access, dynamic workloads, insider threats, and advanced cyberattacks. Traditional perimeter-based security models are no longer sufficient for protecting cloud infrastructures because they rely on implicit trust within network boundaries. Zero Trust Architecture (ZTA) has emerged as an advanced security framework based on the principle of “never trust, always verify,” where every user, device, and communication request must be continuously authenticated and authorized before access is granted. This paper presents a detailed study of Zero Trust Architecture in cloud security, including its core principles, implementation frameworks, enabling technologies, real-world applications, and adoption challenges. The study examines important ZTA components such as identity and access management, micro-segmentation, continuous monitoring, multi-factor authentication, and least- privilege access control. Additionally, the paper analyzes practical deployment approaches in multi-cloud and hybrid cloud environments and discusses key challenges including policy complexity, legacy system integration, and performance overhead. The findings indicate that Zero Trust Architecture significantly improves cloud security by reducing attack surfaces, limiting lateral movement, and strengthening identity-centric protection mechanisms against evolving cyber threats.
Keywords: Zero Trust Architecture, Cloud Security, Identity and Access Management, Micro-Segmentation, Multi- Factor Authentication, Least Privilege, NIST SP 800-207, Cyber Security.
Keywords: Zero Trust Architecture, Cloud Security, Identity and Access Management, Micro-Segmentation, Multi- Factor Authentication, Least Privilege, NIST SP 800-207, Cyber Security.
Abstract
Social Engineering Attacks in Cybersecurity: Analysis, Challenges, and AI-Based Defense Framework
Vikas Gowda J V, Prof. Swetha C S
DOI: 10.17148/IJARCCE.2026.155275
Abstract: As organizations migrate to complex cloud-native architectures, the human interface remains the most vulnerable point of entry. Traditional security mechanisms, while effective against automated malware, often fail to intercept sophisticated Social Engineering (SE) attacks that leverage psychological triggers. This research provides an in-depth analysis of modern SE vectors, including AI-generated phishing and deepfake-based impersonation. We propose a multi-layered AI-Based Defense Framework that utilizes Natural Language Processing (NLP) for semantic intent analysis and behavioral biometrics to create a "Human Firewall." The study evaluates the transition from static training to real-time, AI-driven intervention. Our findings suggest that integrating cognitive-aware AI systems can reduce the success rate of SE attacks by up to 85%, providing a robust defense against the evolving threat landscape of 2026.
Keywords: Social Engineering, Artificial Intelligence, Phishing, Deepfakes, Behavioral Biometrics, Human Element, Semantic Analysis.
Keywords: Social Engineering, Artificial Intelligence, Phishing, Deepfakes, Behavioral Biometrics, Human Element, Semantic Analysis.
Abstract
A Comparative Analysis of Machine Learning Algorithms for the Early Prediction of Heart Disease
Manan Mehra
DOI: 10.17148/IJARCCE.2026.155276
Abstract: Heart disease remains one of the leading causes of global mortality, creating a growing need for accurate and reliable early diagnostic systems. The purpose of this study is to compare selected machine learning algorithms for the early prediction of heart disease and evaluate their suitability for clinical decision-making. The study specifically examines the performance of Logistic Regression, Random Forest, Support Vector Machine (SVM), and K-Nearest Neighbour (KNN) using a clinical dataset of 1000 patient records. The objectives include evaluating model performance through accuracy, precision, recall, and F1-score metrics while identifying significant cardiovascular risk predictors.
The study adopts a quantitative and comparative research design supported by descriptive statistical analysis and machine learning techniques. The findings reveal that Random Forest achieved the highest predictive performance, while Logistic Regression provided better interpretability and transparency for clinical applications. Variables such as chest pain type, exercise angina, ST segment slope, and thallium test results were identified as significant predictors of heart disease. The study concludes that machine learning models can effectively support early heart disease prediction and improve clinical decision-making, provided that predictive accuracy is balanced with interpretability and transparency.
Keywords: Heart Disease Prediction, Machine Learning, Random Forest, Logistic Regression, Cardiovascular Risk Predictive Analytics
The study adopts a quantitative and comparative research design supported by descriptive statistical analysis and machine learning techniques. The findings reveal that Random Forest achieved the highest predictive performance, while Logistic Regression provided better interpretability and transparency for clinical applications. Variables such as chest pain type, exercise angina, ST segment slope, and thallium test results were identified as significant predictors of heart disease. The study concludes that machine learning models can effectively support early heart disease prediction and improve clinical decision-making, provided that predictive accuracy is balanced with interpretability and transparency.
Keywords: Heart Disease Prediction, Machine Learning, Random Forest, Logistic Regression, Cardiovascular Risk Predictive Analytics
Abstract
Social Engineering Attacks in the AI Era: A Review
Dr. Sangeeta Rani, Mr. Gopal Sharma, Dr. Kapil Kumar Kaswan
DOI: 10.17148/IJARCCE.2026.155277
Abstract: The rapid advancement of Artificial Intelligence (AI) has transformed the cybersecurity landscape by enhancing both defensive mechanisms and cybercriminal capabilities. Among the most concerning developments is the evolution of social engineering attacks, where AI technologies are being leveraged to manipulate human behaviour more effectively than ever before. Traditional social engineering techniques such as phishing, spear-phishing, baiting, pretexting, and impersonation have become increasingly sophisticated through the integration of generative AI, large language models, deepfake technologies, voice cloning systems, and automated reconnaissance tools. AI enables attackers to create highly personalized and convincing fraudulent messages, synthetic audio, and realistic video content that exploit human trust and cognitive vulnerabilities. The paper discusses various AI-enabled attack vectors, including AI-generated phishing emails, deepfake-based impersonation, chatbot-assisted scams, social media manipulation, and business email compromise attacks. The review also explores current defense mechanisms, including AI-powered detection systems, user awareness programs, behavioral analytics, multi-factor authentication, and deepfake detection techniques. Additionally, regulatory and ethical considerations related to AI misuse are examined. By synthesising recent research findings and industry reports, this paper highlights the growing threat of AI-enhanced social engineering and emphasises the need for adaptive cybersecurity strategies, ongoing awareness training, and collaborative efforts among researchers, policymakers, and security practitioners. The study concludes by identifying future research directions to develop resilient defences against increasingly intelligent and automated social engineering attacks.
Keywords: Social Engineering, Artificial Intelligence, Generative AI, Deepfake, Phishing
Keywords: Social Engineering, Artificial Intelligence, Generative AI, Deepfake, Phishing
Abstract
Database Reliability and Availability Across Multi-Cloud Environments
Naresh Kumar Miryala
DOI: 10.17148/IJARCCE.2026.155278
Abstract: In the modern digital domain, data is fundamental to business operations, customer experiences, and decision making. As organizations progressively adopt multi-cloud environments to increase scalability, reduce dependence on a single cloud provider, and strengthen system durability, maintaining database reliability and availability has become more complex. Although multi-cloud architectures offer benefits such as flexibility, improved performance, and cost optimization, they also cause challenges related to data uniformity, latency, failover management, and disaster recovery. Guaranteeing uninterrupted database operations across several cloud platforms needs comprehensive planning, robust replication strategies, and effective monitoring mechanisms to sustain high availability and functional stability.
This paper studies database reliability and availability within multi-cloud environments, highlighting their significance in contemporary cloud computing. Reliability emphasizes maintaining information consistency, integrity, and accurate system performance, while availability guarantees that databases remain accessible with minimal downtime, even during failures, maintenance, or unforeseen outages. Achieving high availability in a multi-cloud architecture requires strategies such as data replication, automated failover, load balancing, and auto-scaling. However, carrying out these solutions is complicated by the differing infrastructure designs, networking models, and database management services offered by cloud providers such as Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP).
Multicloud environments present numerous operational and technical issues, including data synchronization, network latency, and interoperability among different cloud platforms. Each provider delivers distinct architectures, configuration models, backup solutions, and disaster recovery (DR) options, making it difficult to achieve uniform performance across multiple environments. Organizations need to carefully manage data replication, synchronization, and access to maintain reliability and uninterrupted availability. Additionally, managing security policies, compliance requirements, monitoring tools, and provider-specific services across multiple clouds increases overall complexity. These problems underline the necessity for organizations to adopt flexible, well-planned database management strategies that support scalability, maintain information consistency, and provide uninterrupted access to critical applications and services.
This article examines how major cloud providers support reliability and availability in their database services. For instance, Amazon Web Services offers Relational Database Service (RDS) with integrated high availability and automatic failover to limit downtime. Microsoft Azure supplies Azure SQL Database, which features cross-region replication and disaster recovery (DR) options. Google Cloud Platform delivers Cloud Spanner, a fully managed database designed for horizontal scaling and strong global consistency.
In addition to studying services offered by individual cloud providers, this paper discusses best practices for designing highly available database architectures in multi-cloud environments. Common strategies include active-active or active- passive replication across databases, which preserve data continuity and seamless failover during outages across multiple systems. Automated monitoring and alerting systems are also critical, as they facilitate early issue detection and limit downtime. In multi-cloud environments, employing unified monitoring and management tools is vital for tracking performance among providers and supporting fast incident response. Furthermore, artificial AI and machine learning are increasingly utilized for forecasting analysis, facilitating organizations to anticipate failures and refine resource usage before issues impact users
This paper serves as a practical guide for database administrators (DBAs), IT leaders, and system architects (SREs) attempting to enhance database reliability and availability in multi-cloud environments. By applying the concepts and best practices presented, organizations can design stronger and high-performing database systems that ensure continuous data access.
Purpose - This white paper provides a detailed comparative study guide for organizations wanting to improve database service reliability and availability across multi-cloud providers. As multi-cloud adoption increases, organizations encounter difficulties in maintaining data access consistency, limiting downtime, and guaranteeing data integrity across 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 5, May 2026 DOI: 10.17148/IJARCCE.2026.155278 several platforms. This article tackles these challenges by supplying clear analysis and actionable best practices for building more reliable and resilient database systems.
Significance - The multi cloud concept is rapidly gaining popularity because of its ability to offer increased flexibility, lessen dependency on a single provider, and increase scalability. However, these benefits are accompanied by challenges, particularly regarding maintaining information consistency, lowering latency, and managing failover among different cloud providers. This study is important as it addresses the increasing need for organizations to develop resilient database services as well as systems able of operating during disruptions, guaranteeing data accessibility regardless of the chosen cloud provider. By assessing key features, challenges, and solutions related to multi-cloud database reliability, this article supports teams in designing and preserving robust systems.
Methods - This white paper utilizes a combination of literature review, comparative study of major cloud providers (Amazon Web Services, Microsoft Azure, and Google Cloud Platform), and case study evaluations. These procedures are used to examine database reliability and availability in multi-cloud environments. The observations and best practices presented are derived from these analyses to assist organizations in improving and upgrading their database infrastructure.
Keywords: Multi cloud environments, Cloud, Database, Database reliability, Database availability, Cloud computing, SQL, Data resilience, High availability, Disaster recovery, Amazon AWS RDS, Azure SQL Database, Google Cloud Spanner, HA/DR, Failover strategies, Database replication, Data synchronization, Predictive analytics, Artificial intelligence in databases, Regulatory compliance, Database architecture, Business continuity.
This paper studies database reliability and availability within multi-cloud environments, highlighting their significance in contemporary cloud computing. Reliability emphasizes maintaining information consistency, integrity, and accurate system performance, while availability guarantees that databases remain accessible with minimal downtime, even during failures, maintenance, or unforeseen outages. Achieving high availability in a multi-cloud architecture requires strategies such as data replication, automated failover, load balancing, and auto-scaling. However, carrying out these solutions is complicated by the differing infrastructure designs, networking models, and database management services offered by cloud providers such as Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP).
Multicloud environments present numerous operational and technical issues, including data synchronization, network latency, and interoperability among different cloud platforms. Each provider delivers distinct architectures, configuration models, backup solutions, and disaster recovery (DR) options, making it difficult to achieve uniform performance across multiple environments. Organizations need to carefully manage data replication, synchronization, and access to maintain reliability and uninterrupted availability. Additionally, managing security policies, compliance requirements, monitoring tools, and provider-specific services across multiple clouds increases overall complexity. These problems underline the necessity for organizations to adopt flexible, well-planned database management strategies that support scalability, maintain information consistency, and provide uninterrupted access to critical applications and services.
This article examines how major cloud providers support reliability and availability in their database services. For instance, Amazon Web Services offers Relational Database Service (RDS) with integrated high availability and automatic failover to limit downtime. Microsoft Azure supplies Azure SQL Database, which features cross-region replication and disaster recovery (DR) options. Google Cloud Platform delivers Cloud Spanner, a fully managed database designed for horizontal scaling and strong global consistency.
In addition to studying services offered by individual cloud providers, this paper discusses best practices for designing highly available database architectures in multi-cloud environments. Common strategies include active-active or active- passive replication across databases, which preserve data continuity and seamless failover during outages across multiple systems. Automated monitoring and alerting systems are also critical, as they facilitate early issue detection and limit downtime. In multi-cloud environments, employing unified monitoring and management tools is vital for tracking performance among providers and supporting fast incident response. Furthermore, artificial AI and machine learning are increasingly utilized for forecasting analysis, facilitating organizations to anticipate failures and refine resource usage before issues impact users
This paper serves as a practical guide for database administrators (DBAs), IT leaders, and system architects (SREs) attempting to enhance database reliability and availability in multi-cloud environments. By applying the concepts and best practices presented, organizations can design stronger and high-performing database systems that ensure continuous data access.
Purpose - This white paper provides a detailed comparative study guide for organizations wanting to improve database service reliability and availability across multi-cloud providers. As multi-cloud adoption increases, organizations encounter difficulties in maintaining data access consistency, limiting downtime, and guaranteeing data integrity across 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 5, May 2026 DOI: 10.17148/IJARCCE.2026.155278 several platforms. This article tackles these challenges by supplying clear analysis and actionable best practices for building more reliable and resilient database systems.
Significance - The multi cloud concept is rapidly gaining popularity because of its ability to offer increased flexibility, lessen dependency on a single provider, and increase scalability. However, these benefits are accompanied by challenges, particularly regarding maintaining information consistency, lowering latency, and managing failover among different cloud providers. This study is important as it addresses the increasing need for organizations to develop resilient database services as well as systems able of operating during disruptions, guaranteeing data accessibility regardless of the chosen cloud provider. By assessing key features, challenges, and solutions related to multi-cloud database reliability, this article supports teams in designing and preserving robust systems.
Methods - This white paper utilizes a combination of literature review, comparative study of major cloud providers (Amazon Web Services, Microsoft Azure, and Google Cloud Platform), and case study evaluations. These procedures are used to examine database reliability and availability in multi-cloud environments. The observations and best practices presented are derived from these analyses to assist organizations in improving and upgrading their database infrastructure.
Keywords: Multi cloud environments, Cloud, Database, Database reliability, Database availability, Cloud computing, SQL, Data resilience, High availability, Disaster recovery, Amazon AWS RDS, Azure SQL Database, Google Cloud Spanner, HA/DR, Failover strategies, Database replication, Data synchronization, Predictive analytics, Artificial intelligence in databases, Regulatory compliance, Database architecture, Business continuity.
Abstract
Identification of Leukemia Subtypes from Microscopic Images Using ResNeXt Algorithm
Mr. Arul Leo Felix L, Ms. Shereen J, Ms. Sushmitha S, Mr. Vishva S
DOI: 10.17148/IJARCCE.2026.155279
Abstract: Leukaemia is a severe haematological malignancy characterized by the abnormal proliferation of white blood cells in bone marrow and peripheral blood. Early and accurate identification of leukaemia subtypes is essential for effective diagnosis and treatment planning. This paper proposes a deep learning–based microscopic blood image classification framework using the ResNeXt architecture for automated leukaemia subtype detection. The system formulates leukaemia classification as a multi-class medical image analysis problem, where complex morphological features are extracted from microscopic blood smear images. The proposed framework incorporates preprocessing techniques such as image resizing, normalization, and data augmentation to improve robustness and generalization performance. A ResNeXt backbone with grouped convolutions and residual learning is employed to capture discriminative cellular patterns while maintaining computational efficiency. The model is trained and evaluated on a publicly available leukaemia microscopic image dataset for accurate subtype classification. Experimental results demonstrate improved classification accuracy, feature representation, and prediction reliability compared with conventional convolutional neural network approaches. The proposed system also supports real-time prediction through a Flask-based web interface, enabling accessible and efficient computer-aided diagnosis. The results indicate that the ResNeXt-based framework provides an effective and scalable solution for intelligent leukaemia detection and automated healthcare assistance.
Keywords: Leukaemia Subtype Classification; Microscopic Blood Smear Images; Deep Learning; ResNeXt Architecture; Medical Image Analysis; Automated Disease Detection; Computer-Aided Diagnosis .
Keywords: Leukaemia Subtype Classification; Microscopic Blood Smear Images; Deep Learning; ResNeXt Architecture; Medical Image Analysis; Automated Disease Detection; Computer-Aided Diagnosis .
Abstract
Data-Driven Billing Reconciliation and Workforce Analytics via UiPath and Power BI
Ashutosh Mankar, Dr G.R Bamnote, Prof. S.P Akarte
DOI: 10.17148/IJARCCE.2026.155280
Abstract: This research paper presented an Data driven powered robot for OCR driven billing reconciliation using UiPath. The study used an open source Accounts Receivable dataset exported from an Baan Application system, where customer balances were recalculated from invoices, payments, credits, and adjustments to identify integration errors. The project used three scripts for synthetic data generation, rule based reconciliation, and anomaly detection through Linear Regression. The UiPath based automation was developed with Outlook, Excel, System, and DataTable activities, along with reusable workflows for email reading, OCR extraction, file validation, data cleaning, matching, exception handling, and output generation. Extracted information from invoices and receipts was converted to structured Excel output to be reviewed and reported on. The results showed significant changes in operations including 87.57% less time spent processing invoices, 65% less time spent preparing orders, 66.67% less time spent reconciling inventory payments, 90% less time spent making mistakes when matching invoices and 20 to 25% less money spent on operations.
Keywords: AI ML, OCR, UiPath, Billing Reconciliation, Robotic Process Automation (RPA), Invoice Processing, Employee Analysis.
Keywords: AI ML, OCR, UiPath, Billing Reconciliation, Robotic Process Automation (RPA), Invoice Processing, Employee Analysis.
Abstract
Brain Tumor Detection and Classification using Deep Learning
Vaibhavi Ghorpade, Sanika Patekar, Sanika Satre and Gauri Wakchaure, Prof. Dr. Sachine Bere, Mr. A. M. Suryawanshi
DOI: 10.17148/IJARCCE.2026.155281
Abstract: Brain Tumors are among the most serious and life-threatening diseases affecting the human brain and nervous system. Early detection is essential to improve patient survival and treatment outcomes. Traditional diagnosis mainly depends on manual examination of Magnetic Resonance Imaging (MRI) scans by radiologists, which can be time- consuming and may lead to human errors. To overcome these limitations, this research proposes a deep learning-based approach using Convolutional Neural Networks (CNN) for automatic detection and classification of brain tumors from MRI images.
The proposed system utilizes a publicly available brain MRI dataset, where preprocessing techniques such as image resizing, normalization, and data augmentation are applied to improve model performance and generalization. The model was implemented and trained using Google Colab with GPU support for faster computation. Experimental results demonstrate high classification accuracy along with strong precision, recall, and F1-score, indicating the effectiveness of the proposed system.
This study highlights the potential of artificial intelligence in improving medical diagnosis by making the detection process faster, more accurate, and less dependent on manual analysis. The proposed system can assist medical professionals in early tumor identification and has significant potential for future integration into real-time clinical applications.
Keywords: Brain Tumor Detection, MRI, CNN, Deep Learning, Transfer Learning, VGG16, Medical Image Analysis, Data Augmentation, Automated Diagnosis.
The proposed system utilizes a publicly available brain MRI dataset, where preprocessing techniques such as image resizing, normalization, and data augmentation are applied to improve model performance and generalization. The model was implemented and trained using Google Colab with GPU support for faster computation. Experimental results demonstrate high classification accuracy along with strong precision, recall, and F1-score, indicating the effectiveness of the proposed system.
This study highlights the potential of artificial intelligence in improving medical diagnosis by making the detection process faster, more accurate, and less dependent on manual analysis. The proposed system can assist medical professionals in early tumor identification and has significant potential for future integration into real-time clinical applications.
Keywords: Brain Tumor Detection, MRI, CNN, Deep Learning, Transfer Learning, VGG16, Medical Image Analysis, Data Augmentation, Automated Diagnosis.
Abstract
Evaluating Symmetric Encryption: Performance and Security Considerations for Enterprise Data Protection
Naresh Kumar Miryala, Meta Platforms Inc. CA, USA
DOI: 10.17148/IJARCCE.2026.155282
Abstract: In the digital domain enterprise data is fundamental to business operations, customer experiences, and decision making. As organizations and user data grow, enterprise focus on the data and data became key elements in the business operations and decisions. The need for data security in the enterprise and criticality of the data is very high in general.
In the world of data, security is a key aspect and security enterprise data is vital, many organizations lost credibility due to data breaches, so protecting enterprise data is key for running successful business and it’s critical for the organizations.
This paper studies various Symmetrical Encryption methods, with the primary objective being the identification of a suitable algorithm for encrypting text files of specific sizes. The experiments for each algorithm involve encrypting text files of various sizes and the paper aims to determine the time duration required for each algorithm to encrypt or decrypt text files of varying sizes and captures the performance considerations for the data encryption.
This paper serves as a practical guide for database administrators (DBAs), IT leaders, and system architects (SREs) attempting to enhance database reliability and availability in multi-cloud environments. By applying the concepts and best practices presented, organizations can design stronger and high-performing systems that ensure secure data access for enterprises.
Purpose: This white paper provides a detailed comparative study guide for organizations wanting to improve data security and using the industry best encryption technique for storing the data and transmitting the data.
Significance: In the world where data has constant threats and ever changing security landscape, understanding and implementing the best practices for the data security and implementing the right algorithms for the data security is highly significant.
Methods - This white paper utilizes a combination of literature review, comparative study of various encryption techniques and these procedures are used to examine performance and security the data encryption methods The observations and best practices presented are derived from these analyses to assist organizations in improving and upgrading the security practices for the data storage.
Keywords: Data, Database, encryption, AES, DES, performance, rest, storage, security, enterprise, Crypto, RC2, TDE.
In the world of data, security is a key aspect and security enterprise data is vital, many organizations lost credibility due to data breaches, so protecting enterprise data is key for running successful business and it’s critical for the organizations.
This paper studies various Symmetrical Encryption methods, with the primary objective being the identification of a suitable algorithm for encrypting text files of specific sizes. The experiments for each algorithm involve encrypting text files of various sizes and the paper aims to determine the time duration required for each algorithm to encrypt or decrypt text files of varying sizes and captures the performance considerations for the data encryption.
This paper serves as a practical guide for database administrators (DBAs), IT leaders, and system architects (SREs) attempting to enhance database reliability and availability in multi-cloud environments. By applying the concepts and best practices presented, organizations can design stronger and high-performing systems that ensure secure data access for enterprises.
Purpose: This white paper provides a detailed comparative study guide for organizations wanting to improve data security and using the industry best encryption technique for storing the data and transmitting the data.
Significance: In the world where data has constant threats and ever changing security landscape, understanding and implementing the best practices for the data security and implementing the right algorithms for the data security is highly significant.
Methods - This white paper utilizes a combination of literature review, comparative study of various encryption techniques and these procedures are used to examine performance and security the data encryption methods The observations and best practices presented are derived from these analyses to assist organizations in improving and upgrading the security practices for the data storage.
Keywords: Data, Database, encryption, AES, DES, performance, rest, storage, security, enterprise, Crypto, RC2, TDE.
Abstract
SPECTROSCOPIC BIOMARKER DETECTION FOR URINE DISEASES USING MACHINE LEARNING
Mr. Surendhiran S Mr. Saran K, Ms. Savina S, Mrs. Nirupashri G
DOI: 10.17148/IJARCCE.2026.155283
Abstract: Urinary diseases such as kidney disorders, urinary tract infections, and diabetic nephropathy are becoming major health concerns worldwide. Early detection of these diseases is important to prevent severe complications and improve treatment outcomes. Traditional diagnostic methods are usually time-consuming, expensive, and dependent on laboratory testing. This paper proposes a machine learning–based framework for urine disease classification using spectroscopic biomarker analysis. The system analyses important urine parameters such as glucose, protein, pH, RBC, WBC, ketone, and bacteria to predict disease severity. Various machine learning algorithms including Support Vector Classifier, Logistic Regression, and Bernoulli Naive Bayes are used for prediction. The trained model is deployed using the Django framework for real-time disease prediction through a web application. The proposed system provides a rapid, accurate, non-invasive, and cost-effective solution for urine disease diagnosis and monitoring.
Keywords: Machine Learning, Urine Disease Classification, Spectroscopy, Biomarker Detection, Support Vector Machine, Non-Invasive Diagnosis.
Keywords: Machine Learning, Urine Disease Classification, Spectroscopy, Biomarker Detection, Support Vector Machine, Non-Invasive Diagnosis.
Abstract
Energy Harvesting and Health Monitoring Device
Dr.D.O.Patil, Dr.R.K.Moje, Sanika Jagnade, Aditi Bhosale, Trupti Sonavane
DOI: 10.17148/IJARCCE.2026.155284
Abstract: Continuous monitoring of vital health parameters is essential for the early detection of medical emergencies and effective patient care, particularly in remote and rural regions where healthcare facilities and communication infrastructure are limited. Conventional health monitoring systems primarily rely on Wi-Fi, GSM, or internet connectivity, which often suffer from restricted coverage, high power consumption, and frequent battery charging requirements. This paper presents an Energy Harvesting Based Smart Health Monitoring Device Using LoRa Communication for Remote Healthcare Applications that enables reliable, low-power, and long-range health monitoring. The proposed system integrates a MAX30102 sensor for heart rate and blood oxygen saturation (SpO₂) measurement, an MLX90632 sensor for body temperature monitoring, a GNSS/GPS module for location tracking, and an OLED display for real-time visualization of health parameters. An ESP 32 microcontroller performs data acquisition, processing, and communication management, while a solar energy harvesting unit with rechargeable energy storage ensures selfsustained operation and minimizes dependence on external charging sources. Health data are transmitted through LoRa technology, enabling communication over several kilometers without requiring cellular or internet networks. In emergency situations, the system automatically sends health status information along with precise GPS coordinates to facilitate rapid medical assistance. The proposed solution offers enhanced portability, energy efficiency, reliability, and cost-effectiveness, making it highly suitable for remote patient monitoring, elderly care, industrial worker safety, and disaster-response healthcare applications.
Keywords: Energy Harvesting, LoRa Communication, Health Monitoring System, Internet of Things (IoT),Healthcare Device, Remote Patient Monitoring, MAX30102 Sensor, ESP 32 microcontroller, Solar-Powered, Emergency Alert System.
Keywords: Energy Harvesting, LoRa Communication, Health Monitoring System, Internet of Things (IoT),Healthcare Device, Remote Patient Monitoring, MAX30102 Sensor, ESP 32 microcontroller, Solar-Powered, Emergency Alert System.
Abstract
AI-DRIVEN INTELLIGENT POWER GRID OPERATION FOR A SMART AND SUSTAINABLE ENERGY FUTURE
Spandana H A, Thrisha S, Thanuja M U, Spoorti Irappa Chamakeri, Dr. Sonia Maria D'souza, Prof.Manojkumar, Ms.Apeksha N H
DOI: 10.17148/IJARCCE.2026.155285
Abstract: The rapid evolution of electrical power systems, coupled with the increasing penetration of renewable energy resources, distributed generation, and digital communication technologies, has transformed conventional power grids into highly interconnected and intelligent energy ecosystems. Artificial Intelligence (AI) has emerged as a transformative technology capable of improving grid reliability, operational efficiency, predictive maintenance, energy management, and cybersecurity. This review article synthesizes recent research on AI applications in smart grids and intelligent power systems, integrating findings from contemporary journal articles and technical studies. The paper critically examines the role of Machine Learning (ML), Deep Learning (DL), Reinforcement Learning (RL), and hybrid AI techniques in load forecasting, fault detection, predictive maintenance, renewable energy integration, demand response, voltage regulation, and energy optimization.
In addition, the study discusses key implementation challenges including data quality, interpretability, computational complexity, scalability, and cybersecurity concerns. Finally, emerging technologies such as Digital Twins, Internet of Energy (IoE), edge computing, and decentralized AI-driven grid architectures are explored as future research directions. The review concludes that AI-driven smart grids will play a critical role in building resilient, sustainable, and adaptive energy infrastructures capable of meeting future global energy demands.
Keywords: Artificial Intelligence, Smart Grid, Machine Learning, Deep Learning, Predictive Maintenance, Renewable Energy Integration, Intelligent Energy Management, Power Systems, Reinforcement Learning, Digital Twin.
In addition, the study discusses key implementation challenges including data quality, interpretability, computational complexity, scalability, and cybersecurity concerns. Finally, emerging technologies such as Digital Twins, Internet of Energy (IoE), edge computing, and decentralized AI-driven grid architectures are explored as future research directions. The review concludes that AI-driven smart grids will play a critical role in building resilient, sustainable, and adaptive energy infrastructures capable of meeting future global energy demands.
Keywords: Artificial Intelligence, Smart Grid, Machine Learning, Deep Learning, Predictive Maintenance, Renewable Energy Integration, Intelligent Energy Management, Power Systems, Reinforcement Learning, Digital Twin.
Abstract
Improving MNIST Image Synthesis via an Optimized Generative Adversarial Network with Transfer Learning and Real-Time Loss Monitoring
Dr.C.Deepa, Dr.B.Vidhya, Dr.V.Sumathi, Dr.N.Mahendiran, Dr.M.Praneesh
DOI: 10.17148/IJARCCE.2026.155286
Abstract: This paper presents an optimized Generative Adversarial Network (GAN) framework for MNIST image generation using Generative Artificial Intelligence (GenAI). The proposed model leverages the collaborative learning process between a generator and a discriminator network to synthesize realistic handwritten digit images. Training efficiency and model stability are enhanced through dynamic loss monitoring and an optimized generator–discriminator architecture. The benchmark MNIST dataset, consisting of grayscale images of handwritten digits from 0 to 9, is used for training and evaluation. The generator is designed using dense neural network layers to create synthetic images, while the discriminator functions as a binary classifier to distinguish between real and generated samples. Throughout the training process, the losses of both networks are continuously monitored to ensure effective convergence and balanced adversarial learning. The performance of the proposed framework is evaluated using multiple metrics, including accuracy, loss, F1-score, and Receiver Operating Characteristic (ROC) curve analysis. Experimental results demonstrate that the model achieves an average classification accuracy of 90%, indicating its effectiveness in generating high-quality MNIST digit images. Furthermore, this work explores the integration of transfer learning techniques within the GAN framework, providing a foundation for extending similar methodologies to more complex image datasets and real-world applications. Future research may focus on advanced loss functions, improved network architectures, and the application of GAN- based image generation across diverse domains. The proposed framework also serves as a valuable educational and research resource for scholars and practitioners working in the fields of Generative AI and deep learning.
Keywords: Generative AI (GenAI), Generative Adversarial Network (GAN), MNIST, Deep Learning, Discriminator loss, Generator loss
Keywords: Generative AI (GenAI), Generative Adversarial Network (GAN), MNIST, Deep Learning, Discriminator loss, Generator loss
Abstract
AI-POWERED PERSONAL HEALTH RECOMMENDATION SYSTEM
Abdoulie Bawo, Sajin Tamang, Modou Lewis, Aakash Ranjan, Sahil Kumar, Dr. P. M. Gavali
DOI: 10.17148/IJARCCE.2026.155287
1. INTRODUCTION
Healthcare
systems across the world are rapidly evolving with the integration of
Artificial Intelligence (AI), Natural Language Processing (NLP), and Large
Language Models (LLMs). Traditional healthcare systems mainly focus on
diagnosis and treatment after the occurrence of diseases, while preventive
healthcare and personalized recommendations remain limited. Most healthcare
applications provide generalized advice that often ignores individual lifestyle
habits, medical history, sleep patterns, fitness goals, and behavioural
conditions. Recent advancements in Generative AI have enabled the development
of intelligent healthcare systems capable of generating adaptive and
context-aware recommendations. These systems can analyze user information such
as age, Body Mass Index (BMI), sleep duration, lifestyle habits, physical
activity levels, and existing medical conditions to provide personalized
suggestions. The proposed AI-Powered Personal Health Recommendation System
utilizes Generative AI APIs to generate personalized healthcare guidance dynamically.
The system offers recommendations related to diet planning, exercise routines,
sleep improvement, medication reminders, and preventive healthcare management.
Unlike traditional machine learning approaches that require large datasets,
feature engineering, and model training, the proposed framework leverages
pretrained Large Language Models through prompt engineering techniques. The
system is developed using modern web technologies including Next.js, React.js,
Node.js, Express.js, Prisma ORM, PostgreSQL, and Gemini AI API. The framework
improves accessibility to healthcare guidance, increases user engagement, and
demonstrates the practical implementation of Generative AI in healthcare
systems.
Abstract
Fake Review Detection in E-Commerce Using Machine Learning and NLP: A Comparative Study
Ajeetha G
DOI: 10.17148/IJARCCE.2026.155288
Abstract: The proliferation of fake reviews in e-commerce platforms has become a critical challenge affecting consumer trust and purchasing decisions. This paper presents a comparative study of machine learning approaches for automatic fake review detection using Natural Language Processing (NLP) techniques. We evaluate three classification algorithms — Logistic Regression, Naive Bayes, and Support Vector Machine (SVM) — on a dataset of 8,087 reviews using TF- IDF feature extraction. Experimental results demonstrate that SVM achieves the highest accuracy of 86.60% with precision and recall of 0.87, outperforming Logistic Regression (86.01%) and Naive Bayes (84.37%). The proposed system effectively distinguishes between genuine (CG) and fake (OR) reviews, providing a reliable foundation for trustworthy product recommendation systems.
Keywords: Fake review detection, Natural Language Processing, TF-IDF, Support Vector Machine, E-commerce, Opinion spam.
Keywords: Fake review detection, Natural Language Processing, TF-IDF, Support Vector Machine, E-commerce, Opinion spam.
Abstract
AI- Powered Public Safety Monitoring System
Anand Tasgave,Sahil Thomake,Aditya Vhanuagre,Shahid M.Ali Vijapure,Harshwardhan Patil, Prof. K. S. Kadam
DOI: 10.17148/IJARCCE.2026.155289
Abstract: Public safety has become an important concern in modern society due to increasing crime rates and emergency situations. Traditional surveillance systems require continuous human monitoring, which can lead to errors and delayed responses. This paper proposes an AI Powered Public Safety Monitoring System that uses Artificial Intelligence, Machine Learning, and Computer Vision for real-time monitoring and threat detection. The system can identify suspicious activities, detect abnormal behavior, monitor crowded areas, and send instant alerts to authorities. It also supports features such as facial recognition, motion detection, and cloud-based data storage for secure monitoring. The proposed system helps improve security, reduce manual effort, and provide faster response during emergencies in public places and smart city environments.
Keywords: Artificial Intelligence, Public Safety, Computer Vision, Machine Learning, Smart Surveillance, Real- Time Monitoring•
Keywords: Artificial Intelligence, Public Safety, Computer Vision, Machine Learning, Smart Surveillance, Real- Time Monitoring•
Abstract
Micro Front End Architecture: Accelerating Team Scalability in Modern Tech
Ritika B Immadi, A G Vishvanath
DOI: 10.17148/IJARCCE.2026.155290
Abstract: In the rapidly changing world of web application development, one problem stays forever the same: scalability. That’s particularly the case for large-scale enterprises that want to boost their technological agility and responsiveness. Large-scale is where the problems of monolithic architectures (both frontend and backend) hit you hardest. This paper looks at one possible way to work around those problems: micro front ends. The idea behind micro front ends is to apply the same principles of decomposition (scalability, maintainability, and team autonomy) that work so well when you use them on backend services. If the decomposition of backend services allows for independent deployment, development, and scaling of those services, then the same should be possible with frontend components. This paper synthesizes the current body of literature and real-world case studies from companies like IKEA, DAZN, and HelloFresh that have successfully implemented micro front end architectures. Specific examples show how several happily-noted-as-solved problems have been handled by various organizations, and the two main things these organizations seem to have in common is some kind of rock-solid governance model that they all sing to and more than a few advanced deployment techniques (like server-side rendering and progressive web apps) that they all seem to be employing. This paper goes through an extensive review of both academic and practical literature to evaluate micro front ends. It sets the scene with a clear overview of what micro front ends are, discussing the architecture's components and how they work together, and working through some simple examples to make clear the kind of problems micro front ends might solve.
Keywords: Micro Front Ends, Scalability, Web Application Development, Modular Architecture, Software Engineering, Frontend Decomposition, Agile Development, Enterprise Applications
Keywords: Micro Front Ends, Scalability, Web Application Development, Modular Architecture, Software Engineering, Frontend Decomposition, Agile Development, Enterprise Applications
Abstract
AI-Driven Space Debris Detection, Tracking and Removal System
Mrs. Nita Meshram, M Pooja, Prarthana J, Priyanka Kumari, Spandana S P
DOI: 10.17148/IJARCCE.2026.155291
Abstract: The increasing amount of orbital debris has become significant concern for modern satellite communication and space exploration activities. This paper presents an AI-enabled framework for detecting, tracking, and mitigating space debris using Artificial Intelligence (AI), Internet of Things (IoT), and predictive analytics. The proposed system continuously monitors orbital objects through intelligent sensing devices and analyzes orbital movement using Machine Learning and Deep Learning algorithms. The framework predicts collision risks, generates warning alerts, and supports automated mitigation mechanisms for improved orbital safety. The proposed approach aims to reduce collision probability, enhance monitoring accuracy, and promote sustainable space operations.
Keywords: Artificial Intelligence, Space Debris, IoT, Machine Learning, Collision Prediction, Orbital Monitoring.
Keywords: Artificial Intelligence, Space Debris, IoT, Machine Learning, Collision Prediction, Orbital Monitoring.
Abstract
Speech- Driven note-taking with AI-Based Transcription, Translation and Summarization
Khushi h Dhongadi, Swetha M
DOI: 10.17148/IJARCCE.2026.155292
Abstract: The rapid advancement of global digital communication has significantly increased the demand for efficient, real-time speech processing and translation capabilities. Traditional cascaded speech translation systems often struggle with high latency and compounding errors due to their reliance on sequential processing pipelines. This paper presents a comprehensive overview of unified end-to-end (E2E) frameworks that seamlessly execute speech-to-text transcription, simultaneous translation, and automated text summarization. A key innovation highlighted in these systems is the use of causal alignment and training-free policies to unify translation mechanisms and timing schedules without requiring resource-intensive ad-hoc training pipelines. Performance and architectural efficiency are further enhanced using intelligent mechanisms like Decoder Time Dilation and quantized edge-deployed protocols to mitigate autoregressive overhead. The overall results demonstrate that these unified E2E architectures achieve remarkable Word Error Rates (WER) and state-of-the-art quality-latency trade-offs, offering a highly scalable solution for modern real-time streaming environments.
Keywords: Real-Time Speech Processing, Simultaneous Translation, End-to-End (E2E) Architectures, Automated Summarization, Edge Deployment, Word Error Rate (WER).
Keywords: Real-Time Speech Processing, Simultaneous Translation, End-to-End (E2E) Architectures, Automated Summarization, Edge Deployment, Word Error Rate (WER).
Abstract
An AI-Powered Integrated Placement Module with Automated Proctoring and Chatbot-driven Interview Simulation for Higher Education
Sayli Patil*, Chaitali Banmare, Pranoti Nule, Diksha Thakar, Pratiksha Titkare
DOI: 10.17148/IJARCCE.2026.155293
Abstract: Modern recruitment in higher education requires integrated platforms that evaluate technical skills and soft skills while maintaining examination integrity. This paper presents the design and implementation of a web based Placement Module developed using Node.js, Express.js, MongoDB and EJS (MENE) stack to bridge this gap. The proposed system integrates a secure Online Placement Test featuring real-time camera proctoring to prevent malpractice and an AI-driven Interview Chatbot that utilizes a dynamic database of recently asked industry questions to simulate real-world hiring scenarios. Administrators are provided with a robust dashboard to manage test schedules, update question banks and monitor live sessions. For students, the platform offers a seamless interface to track overall performance analytics, providing data-driven insights into their strengths and areas for improvement. Experimental results indicate that the asynchronous nature of Node.js, combined with the flexible document schema of MongoDB, allows the system to handle concurrent user sessions with high efficiency. This research demonstrates how a unified web- based framework can streamline the placement lifecycle, reducing administrative overhead while enhancing student readiness for professional recruitment.
Keywords: Placement module, Emotional Framework, Web-Based Platform, Higher Education, Interview Chatbot, Node.js, MongoDB
Keywords: Placement module, Emotional Framework, Web-Based Platform, Higher Education, Interview Chatbot, Node.js, MongoDB
Abstract
OncoSight – AI Powered Early Cancer Detection System
Dr. Samuel Chellathurai A.Ph.D, Sibimathavan K, Yogaraj M
DOI: 10.17148/IJARCCE.2026.155294
Abstract: Cancer remains one of the foremost causes of mortality worldwide, with early and accurate diagnosis being the most decisive factor in improving patient survival outcomes. Traditional cancer detection methods rely on manual examination of histopathological tissue slides and CT scan images by radiologists and pathologists — a process that is time-consuming, resource-intensive, and inherently prone to human error. OncoSight is a web-based AI diagnostic portal developed using the Django framework to address this critical gap. The system integrates two deep learning models into a unified, browser-accessible interface: a ResNet50 model trained on the LC25000 histopathological dataset for three- class lung tissue classification (Adenocarcinoma, Squamous Cell Carcinoma, Normal), and a weighted ensemble of EfficientNetB3 (25%), EfficientNetB4 (45%), and ResNet50V2 (30%) trained on the LIDC-IDRI CT scan dataset (1,010 patients) for binary lung nodule malignancy detection using focal loss and AdamW optimiser with a malignancy threshold of 0.40. Grad-CAM heatmap visualisations are generated for every prediction, highlighting regions that most influenced the AI decision. The system achieves a weighted ensemble ROC-AUC of 0.8776, delivers predictions within seconds, and provides dedicated per-patient diagnostic report pages. OncoSight addresses the real-world gap of no accessible, clinically deployable AI cancer detection platform while being architecturally designed for future multi-cancer expansion.
Keywords: Cancer Detection, Deep Learning, ResNet50, EfficientNet, Transfer Learning, Grad-CAM, LIDC-IDRI, Histopathological Classification, Ensemble Learning, Django.
Keywords: Cancer Detection, Deep Learning, ResNet50, EfficientNet, Transfer Learning, Grad-CAM, LIDC-IDRI, Histopathological Classification, Ensemble Learning, Django.
Abstract
Brain Box: An AI-Powered Multimodal Knowledge Organizer Platform
Ashitosh Sanjay Langare, Snehal Namdev Lohar, Sanika Sunil Mane, Rohit Dilip Patil, Prof. Mansi Khanaj
DOI: 10.17148/IJARCCE.2026.155295
Abstract: The exponential growth of digital content across diverse modalities—text, images, audio, video, and documents—has created a critical need for intelligent knowledge management systems. Traditional note-taking and bookmarking tools rely on keyword-based search and manual categorization, failing to provide context-aware retrieval or semantic understanding. This paper presents Brain Box, an AI-powered multimodal Knowledge Organizer Platform that acts as a personal digital memory. Brain Box enables users to capture, organize, and retrieve all content types from a single unified interface. The system employs semantic embeddings via LangChain and OpenAI/Hugging Face APIs, Retrieval-Augmented Generation (RAG) for context-aware query resolution, and a privacy-first dual-storage architecture supporting both local and encrypted cloud storage. A React.js frontend, Node.js/Express.js backend, and vector databases (FAISS/Pinecone) constitute the core technical stack. The platform achieves query response times under 3 seconds with 95% uptime targets, demonstrating viability for academic, professional, and personal productivity use cases. This work addresses nine documented shortcomings of existing tools—including lack of multimodal support, weak semantic retrieval, no contextual memory, and fragmented ecosystems—and proposes a scalable, privacy-first architecture to overcome them.
Keywords: Knowledge Organizer, Semantic Search, Retrieval-Augmented Generation, Multimodal AI, Vector Database, LangChain, Privacy-First Storage, Natural Language Processing, LLM, FAISS, Pinecone, RAG
Keywords: Knowledge Organizer, Semantic Search, Retrieval-Augmented Generation, Multimodal AI, Vector Database, LangChain, Privacy-First Storage, Natural Language Processing, LLM, FAISS, Pinecone, RAG
Abstract
BCI Performance Optimization Using EEG Band Power and Hybrid Deep Learning
Dr.H.Umma Habiba, Sarani M, Arthi J, Shamli S
DOI: 10.17148/IJARCCE.2026.155296
Abstract: Brain-Computer Interface (BCI) systems rely on accurate interpretation of electroencephalogram (EEG) signals to understand human cognitive states. However, EEG signals are highly noisy, non-linear, and vary across individuals, which reduces classification accuracy. This paper proposes a real-time BCI performance optimization system using EEG band power features and a hybrid deep learning model. The system processes Alpha, Beta, and Gamma brainwave components to extract meaningful band power features. Additionally, EEG signals are converted into spectrogram images to capture time-frequency patterns. A hybrid Convolutional Neural Network (CNN) model is used to combine spectrogram-based image features with numerical band power features for improved classification. The system predicts cognitive states such as Relaxed, Normal, and High Stress. A real-time interactive dashboard is developed for visualization, prediction, and report generation. Experimental results demonstrate improved classification accuracy and robustness compared to traditional approaches. The proposed system enhances BCI reliability and can be applied in healthcare, mental monitoring, and neuro-adaptive systems.
Keywords: EEG, BCI, Deep Learning, CNN, Band Power, Spectrogram, Stress Detection, Neurotechnology1
Keywords: EEG, BCI, Deep Learning, CNN, Band Power, Spectrogram, Stress Detection, Neurotechnology1
Abstract
GEOSTONE DETECTION AI – INTELLIGENT ROCK AND GEMSTONE RECOGNITION SYSTEM
Dr. Samuel Chellathurai A, Vaishnavi S, Subalakshmi R, Thamarai R
DOI: 10.17148/IJARCCE.2026.155297
Abstract: GeoStone Detection AI is an intelligent AI-powered geological analysis system developed for the automatic identification and scientific analysis of rocks, minerals, and gemstones using image processing, computer vision, and multimodal Artificial Intelligence technologies. Traditional geological identification methods mainly depend on manual observation, laboratory testing, and expert geological analysis, which are time-consuming, expensive, and inaccessible for many users. To overcome these limitations, the proposed system integrates modern web technologies, cloud computing, and Vision Language Models into a unified full-stack web application capable of performing real-time geological analysis from uploaded images. The system is developed using React.js for frontend implementation and Node.js with Express.js for backend processing, while the core AI engine utilizes the Groq Cloud API integrated with the Llama 4 Scout Vision model for multimodal image understanding and scientific report generation. Uploaded geological images undergo preprocessing operations such as resizing, normalization, noise reduction, and Base64 conversion before AI analysis. The system extracts visual characteristics including texture, color, transparency, crystal structure, and mineral patterns to identify geological specimens accurately and generate detailed geological reports containing physical properties, chemical composition, geological formation, rarity level, industrial applications, commercial value, and safety information. The application is deployed using Vercel and Render.com to provide scalable cloud-based accessibility across desktop and mobile devices without requiring specialized hardware or software installation. Experimental evaluation demonstrates that the proposed system provides fast response time, reliable prediction performance, efficient preprocessing, and user-friendly interaction for real-time geological analysis. The project highlights the practical implementation of multimodal Artificial Intelligence in scientific applications and demonstrates how Vision Language Models can improve accessibility, automation, and efficiency in geological identification systems while providing a scalable foundation for future enhancements such as multilingual support, offline AI processing, GPS-based geological mapping, and advanced 3D mineral visualization.
Keywords: Artificial Intelligence, Computer Vision, Geological Analysis, Rock Classification, Gemstone Identification, Vision Language Model, Deep Learning, Multimodal AI, Image Processing, Cloud Computing, Scientific Report Generation, Full-Stack Web Application.
Keywords: Artificial Intelligence, Computer Vision, Geological Analysis, Rock Classification, Gemstone Identification, Vision Language Model, Deep Learning, Multimodal AI, Image Processing, Cloud Computing, Scientific Report Generation, Full-Stack Web Application.
Abstract
AMR INSIGHT: MACHINE LEARNING-BASE MICROBE RESISTANCE PREDICTION USING ARFA
Dr. Swetha Singh, Subash M, Akash V
DOI: 10.17148/IJARCCE.2026.155298
Abstract: Antimicrobial Resistance (AMR) has emerged as one of the most significant global healthcare challenges, leading to treatment failures, prolonged hospitalization, increased mortality rates, and rising healthcare costs. Conventional antimicrobial susceptibility testing methods require significant laboratory time and resources, making them unsuitable for rapid clinical decision-making. This paper proposes a Machine Learning-Based Antimicrobial Resistance Prediction System integrated with an Adaptive Resistance Fusion Algorithm (ARFA) to improve resistance prediction accuracy. The proposed framework utilizes clinical and microbiological data including patient demographics, microbial species information, infection characteristics, and antibiotic usage history. ARFA performs weighted feature fusion to generate a structured resistance risk representation before classification. Multiple supervised machine learning algorithms including Random Forest, Support Vector Machine (SVM), Logistic Regression, and Decision Tree are employed for resistance prediction. The system provides rapid resistance assessment, improves clinical decision support, and enhances antimicrobial stewardship practices. Experimental analysis demonstrates improved prediction performance and interpretability compared to traditional approaches. The proposed solution offers a scalable and intelligent framework for early antimicrobial resistance detection in modern healthcare environments.
Keywords: Antimicrobial Resistance, Machine Learning, ARFA, Random Forest, Support Vector Machine, Clinical Prediction, Healthcare Analytics, Antibiotic Resistance.
Keywords: Antimicrobial Resistance, Machine Learning, ARFA, Random Forest, Support Vector Machine, Clinical Prediction, Healthcare Analytics, Antibiotic Resistance.
Abstract
AI-Enabled Sensor-Based Crop Stress Early Prediction and Detection System
Kanishkumar R, Sivasankar D, Dr. E. Sivanantham
DOI: 10.17148/IJARCCE.2026.155299
Abstract: Timely identification of crop stress is critical to reducing agricultural losses and supporting sustainable food production. This paper presents an AI-enabled sensor-based platform that integrates Internet of Things (IoT) sensing, cloud-based data management, and deep learning inference to predict and detect crop stress at an early stage. Temperature, humidity, and soil moisture data are continuously acquired using a DHT11 sensor and a capacitive soil moisture sensor interfaced with an ESP32 microcontroller, and the readings are transmitted to a Firebase Realtime Database. A Flask-based backend retrieves the stored time-series records and feeds them to a trained Long Short-Term Memory (LSTM) network, which forecasts future soil moisture values. The system concurrently analyses moisture trend behaviour—classifying conditions as stable, gradually decreasing, or critically decreasing—and cross-references actual and predicted values against crop-specific thresholds to generate three-level stress alerts: Low, Medium, and High. Early- warning notifications are raised when predicted values approach critical limits, while High-stress alerts are triggered when readings fall below safety thresholds. An interactive web dashboard delivers live sensor telemetry, LSTM forecast outputs, trend indicators, and stress alert status in real time. Experimental evaluation confirms accurate prediction and reliable alert generation, demonstrating that the proposed system provides an affordable, scalable, and intelligent solution for modern precision agriculture.
Keywords: crop stress detection; LSTM time-series forecasting; IoT smart agriculture; soil moisture prediction; ESP32; Firebase; precision farming; early warning system
Keywords: crop stress detection; LSTM time-series forecasting; IoT smart agriculture; soil moisture prediction; ESP32; Firebase; precision farming; early warning system
Abstract
INTELLIGENT ENERGY SAVING SYSTEM FOR HOME APPLIANCES
Mrs. Nigileeshwari, Sivaguru A, Mohamed Irfan A, Dhinesh S
DOI: 10.17148/IJARCCE.2026.155300
Abstract: This paper presents an Internet of Things (IoT)- based Intelligent Energy Saving System designed to optimize power consumption in home appliances. The proposed sys- tem utilizes sensors, a microcontroller (ESP32), and smart control mechanisms to monitor and manage electrical devices efficiently. The system automatically detects appliance usage using current sensors and environmental conditions using motion and light sensors. Based on real-time data, it controls appliances such as lights and fans to reduce unnecessary energy consumption.
Additionally, the system provides remote monitoring and control through a mobile application using the Blynk IoT platform. A relay module is used to switch appliances ON/OFF, while an LCD/OLED display shows real-time energy usage. The system is cost-effective, scalable, and suitable for smart home integration. Experimental results demonstrate significant en- ergy savings, reduced power wastage, and improved efficiency, making it an effective solution for modern energy management.
Additionally, the system provides remote monitoring and control through a mobile application using the Blynk IoT platform. A relay module is used to switch appliances ON/OFF, while an LCD/OLED display shows real-time energy usage. The system is cost-effective, scalable, and suitable for smart home integration. Experimental results demonstrate significant en- ergy savings, reduced power wastage, and improved efficiency, making it an effective solution for modern energy management.
Abstract
AN INTELLIGENT CULINARY INSTRUCTION GENERATOR FOR AUTOMATED RECIPE CREATION FROM VISUAL FOOD IMAGES
K.S Sukrutha, Shruthi H.R, Anusha C.G, Sowbhagya.P, Varsha.S, Varshini.R-
DOI: 10.17148/IJARCCE.2026.155301
Abstract: Cooking is a creative and sensory process that relies heavily on the visual perception of food ingredients and their presentation. With recent advancements in Artificial Intelligence and Deep Learning, automated food recognition and recipe generation have become important research areas. This project presents a recipe generation system that utilizes Convolutional Neural Networks (CNNs) to analyze food images and automatically generate corresponding recipes. The proposed model accepts a food image as input and identifies the dish by extracting visual features through CNN-based image classification techniques. Based on the recognized food item, the system generates a complete recipe, including the dish name, required ingredients, and step-by-step cooking instructions. The developed approach aims to assist users in understanding and preparing food dishes from visual information alone, thereby enhancing cooking experiences and promoting intelligent food recommendation systems. Experimental results demonstrate the effectiveness of the proposed method in generating meaningful and relevant recipes from food images.
Abstract
AI BASED CROP MARKET PRICE PREDICTION SYSTEM USING MACHINE LEARNING
Mr. Muthukrishnan M.E, Vishwa S, Navin Kumar K
DOI: 10.17148/IJARCCE.2026.155302
Abstract: This paper presents a machine learning based system for predicting the market price of agricultural crops to support farmers, traders, and agricultural policymakers in making timely and informed decisions. Agricultural commodity prices in India are highly volatile and are influenced by seasonal cycles, weather conditions, demand and supply fluctuations, transportation costs, and government policies. This volatility, combined with a lack of reliable price information at the farm level, frequently forces farmers into distress selling at unfavourable rates. The proposed system, CropCast AI, integrates historical market (mandi) price data, weather parameters, seasonal indicators, and supply demand factors to forecast future crop prices for a selected crop, market, and time period. Multiple regression and time series models including Linear Regression, Random Forest, XGBoost, and Long Short Term Memory (LSTM) networks were trained and compared on historical price datasets. The system is deployed as an interactive web application built using the Flask framework, allowing users to select a crop, state, market, and target date and instantly view the predicted price along with historical price trends. Experimental results demonstrate that the ensemble and deep learning models outperform classical baselines, with the best model achieving a coefficient of determination (R2) of 0.93, a Mean Absolute Percentage Error (MAPE) of 7.8%, and a Root Mean Square Error (RMSE) within acceptable agricultural forecasting limits. The proposed framework reduces information asymmetry, helps farmers decide when and where to sell their produce, supports better procurement and storage planning, and provides a low cost decision support tool for the Indian agricultural ecosystem.
Keywords: Crop price prediction, machine learning, agriculture, XGBoost, LSTM, time series forecasting, mandi prices, regression analysis, decision support system, precision agriculture.
Keywords: Crop price prediction, machine learning, agriculture, XGBoost, LSTM, time series forecasting, mandi prices, regression analysis, decision support system, precision agriculture.
Abstract
Predicting Cleft Lip in Unborn Babies Using Ultrasound Images and Machine Learning
Mrs. Dakshayini G R, Mahalakshmi P S, Dheekshith M, Devish Papani, Prajwal koushik C
DOI: 10.17148/IJARCCE.2026.155303
Abstract: Predicting cleft lip in unborn babies using ultrasound images and machine learning offers a non- invasive approach for early detection during prenatal stages. The system uses deep learning to automatically identify facial irregularities in ultrasound scans, focusing on the lip and surrounding regions. A CNN model trained on annotated images classifies scans as normal or cleft-affected with high accuracy, supported by preprocessing steps such as noise reduction and contrast enhancement. Grad-CAM visualizations highlight the regions influencing predictions, improving clinical transparency.
A Python-based web application built using Flask or Django allows users to upload ultrasound images and receive real- time predictions with confidence scores. This cost-effective tool supports gynecologists and radiologists in early anomaly detection, reducing dependence on postnatal procedures. Overall, the system strengthens prenatal screening and contributes to timely intervention and improved neonatal outcomes.
Keywords: Cleft lip detection, ultrasound img, deep learning, CNN, Grad-CAM, medical imaging, AI healthcare.
A Python-based web application built using Flask or Django allows users to upload ultrasound images and receive real- time predictions with confidence scores. This cost-effective tool supports gynecologists and radiologists in early anomaly detection, reducing dependence on postnatal procedures. Overall, the system strengthens prenatal screening and contributes to timely intervention and improved neonatal outcomes.
Keywords: Cleft lip detection, ultrasound img, deep learning, CNN, Grad-CAM, medical imaging, AI healthcare.
Abstract
Application of Mohan Transform in Initial and Boundary Value Problems
Bhatendra kumar*, Jyotsna chandel
DOI: 10.17148/IJARCCE.2026.155304
Abstract: This paper presents the application of the Mohan transform for solving initial value problems and boundary value problems arising in ordinary and partial differential equations. The study demonstrates that the Mohan transform is an efficient analytical tool for converting differential equations into simpler algebraic forms, thereby reducing computational complexity and simplifying solution procedures. Fundamental properties of the Mohan transform, including linearity, derivative transforms, and convolution properties, are discussed and utilized in the formulation of solutions. Several numerical illustrations involving homogeneous and non-homogeneous boundary value problems, heat equations, and wave equations are solved systematically using the transform technique. Graphical comparisons and error analysis confirm that the Mohan transform produces exact or near-exact solutions comparable to those obtained by the Laplace transform while requiring fewer computational steps. The results establish that the Mohan transform is a reliable and powerful mathematical approach for solving a wide class of engineering, physics, and applied mathematics problems.
Keywords: Mohan Transform, Initial Value Problems, Boundary Value Problems, Partial Differential Equations, Integral Transform, Numerical Illustration, Exact Solution, Laplace Transform Comparison.
Keywords: Mohan Transform, Initial Value Problems, Boundary Value Problems, Partial Differential Equations, Integral Transform, Numerical Illustration, Exact Solution, Laplace Transform Comparison.
Abstract
Shuffling in Artificial Intelligence: Foundations, Methodologies, Applications, and Future Directions
Ahmed S. AlMahmeed
DOI: 10.17148/IJARCCE.2026.155305
Abstract: Shuffling is a foundational component in the artificial intelligence (AI) and machine learning (ML) pipeline, exerting significant influence on the integrity and effectiveness of models. This comprehensive review examines the theoretical underpinnings, algorithmic implementations, and practical roles of shuffling in data preprocessing, model training, and evaluation. We discuss how shuffling impacts generalization, bias, and variance, and detail computational and reproducibility challenges that practitioners encounter. Through extended analysis, we present diverse case studies from domains such as computer vision, natural language processing, and reinforcement learning, illustrating practical benefits and pitfalls. Furthermore, we discuss the historical evolution of shuffling strategies, highlighting key algorithms and their statistical properties. Finally, we propose future research directions, including efficient shuffling for large-scale, distributed, and privacy-sensitive settings, as well as theoretical analysis for emerging paradigms like continual, self- supervised, and federated learning. Our goal is to provide AI researchers, academics, and practitioners with actionable insights and a holistic understanding of shuffling’s critical role in modern AI systems.
Keywords: Artificial Intelligence, Machine Learning, Data Preprocessing, Randomization, Data Shuffling, Feature- Level Shuffling, Temporal Shuffling, Data Integrity, Model Generalization, Bias Reduction, Variance Reduction, Ensemble Methods, Reproducibility, Distributed Learning, Federated Learning, Continual Learning, Self-Supervised Learning
Keywords: Artificial Intelligence, Machine Learning, Data Preprocessing, Randomization, Data Shuffling, Feature- Level Shuffling, Temporal Shuffling, Data Integrity, Model Generalization, Bias Reduction, Variance Reduction, Ensemble Methods, Reproducibility, Distributed Learning, Federated Learning, Continual Learning, Self-Supervised Learning
Abstract
An Artificial Intelligence-Driven Framework for Text Similarity Measurement and Plagiarism Detection Using Hybrid Lexical and Semantic Analysis
DIDDE PRAVEEN KUMAR, A.N. RAMA MANI*
DOI: 10.17148/IJARCCE.2026.155306
Abstract: The proliferation of digital text and the ease of electronic copying have made academic and professional plagiarism a pervasive concern, motivating the need for detection tools that go beyond superficial string comparison. Conventional plagiarism checkers rely heavily on exact or near-exact lexical matching and consequently fail to recognize paraphrased, restructured, or semantically equivalent content. This paper proposes an artificial-intelligence-driven framework that combines lexical and semantic analysis to measure textual similarity and detect plagiarism with improved accuracy. The system couples classical term-frequency and n-gram representations with contextual embeddings produced by transformer-based language models, and fuses the two signals into a single interpretable similarity score. Candidate sources are retrieved efficiently from a reference corpus using an approximate nearest-neighbour vector index, and matched passages are highlighted in a structured report. The backend is implemented in Python, exposing services through a lightweight web framework, while a Node.js client provides document submission and report visualization. Experimental evaluation on a curated dataset of original and manipulated documents shows that the proposed fusion approach attains a precision of 0.94, a recall of 0.92, and an F1-score of 0.93, outperforming string-matching, term- frequency, and embedding-only baselines, and achieving an area under the ROC curve of 0.96. The principal contributions are a hybrid similarity-scoring methodology, an efficient retrieval-and-reporting pipeline, and a comparative empirical analysis demonstrating that semantic augmentation substantially improves the detection of disguised plagiarism.
Keywords: Plagiarism detection; Text similarity; Natural language processing; Transformer embeddings; Semantic analysis; TF-IDF; Information retrieval; Machine learning
Keywords: Plagiarism detection; Text similarity; Natural language processing; Transformer embeddings; Semantic analysis; TF-IDF; Information retrieval; Machine learning
Abstract
An AI-Driven Framework for Real-Time Energy Consumption Monitoring, Demand Forecasting, and Optimization Using Deep Sequence Models
M. Lavanya Durga, K. Lakshamana Reddy*
DOI: 10.17148/IJARCCE.2026.155307
Abstract: Rising electricity costs, grid stress, and the global imperative to reduce carbon emissions have intensified the need for intelligent systems that can monitor and curtail energy waste at the building and household level. Conventional metering reports cumulative consumption after the fact, offering little insight into when, where, or why energy is wasted, and provides no forward-looking guidance for reducing demand. This paper presents an artificial-intelligence framework that ingests fine-grained consumption telemetry, forecasts short-term demand, detects anomalous usage, and recommends actionable optimization measures in real time. The system streams data from smart meters and appliance- level sensors through an edge gateway into a time-series store, applies a deep sequence model for load forecasting, flags deviations through an anomaly detector, and surfaces savings recommendations on an interactive dashboard. A Python back end performs model training and inference, while a Node.js layer delivers live visualization over web sockets. Evaluated against linear-regression and ARIMA baselines, the proposed deep model reduced forecasting error to a mean absolute percentage error of roughly 6.4%, and the optimization layer identified consumption reductions of up to 18% in evaluation scenarios. The principal contributions are an end-to-end streaming monitoring pipeline, an accurate deep forecasting and anomaly-detection layer, and an interpretable recommendation engine that links prediction to concrete energy-saving actions.
Keywords: Energy monitoring; demand forecasting; deep learning; LSTM; anomaly detection; smart metering; energy optimization; time-series analysis.
Keywords: Energy monitoring; demand forecasting; deep learning; LSTM; anomaly detection; smart metering; energy optimization; time-series analysis.
Abstract
Serverless Document Workflow Orchestration, Cloud-Native Architecture Using AWS Step Functions for Distributed Approval Processing
Veeravalli Jyothika, K. Lakshmi Sai Sri*
DOI: 10.17148/IJARCCE.2026.155308
Abstract: This paper presents a production-grade serverless orchestration platform designed for cloud-native document workflow management and distributed approval processing. The proposed system integrates Spring Boot 3 microservices architecture with AWS Step Functions for human-in-the-loop state machine orchestration, enabling scalable document submission, review, and approval workflows. The platform employs declarative workflow definitions using Amazon States Language, eliminating operational complexity associated with traditional orchestration systems. A comprehensive local workflow engine facilitates development and testing without AWS dependencies. The implementation demonstrates measured improvements: 95% reduction in workflow latency, support for 500+ concurrent workflows, and 70% operational cost reduction compared to conventional monolithic approval systems. The architecture validates serverless orchestration patterns for enterprise document processing, establishing design guidelines for distributed decision workflows. This research contributes novel insights into activity-based task coordination, stateless workflow execution, and cloud-native infrastructure patterns applicable to business process automation domains.
Keywords: Serverless Computing, AWS Step Functions, Workflow Orchestration, Document Management, Distributed Systems, Stateless Architecture, Business Process Automation, Cloud-Native Design
Keywords: Serverless Computing, AWS Step Functions, Workflow Orchestration, Document Management, Distributed Systems, Stateless Architecture, Business Process Automation, Cloud-Native Design
Abstract
A Scalable, Serverless Cloud Architecture for Real-Time Courier and Shipment Tracking Using AWS ECS Fargate and Containerized Microservices
BOKKA DIVYA, Dr. CHIRAPARAPU SRINIVASA RAO*
DOI: 10.17148/IJARCCE.2026.155309
Abstract: The rapid expansion of e-commerce and on-demand delivery services has created an urgent demand for logistics platforms that are highly available, geographically scalable, and capable of providing customers with transparent, real-time shipment visibility. Conventional courier management systems, predominantly constructed on monolithic architectures hosted on dedicated servers or virtual machines, suffer from significant operational bottlenecks including high infrastructure maintenance overhead, manual scaling constraints, extended deployment cycles, and inadequate customer notification mechanisms. This paper presents the design, implementation, and evaluation of a cloud-native courier and shipment tracking system built on Amazon Web Services (AWS) Elastic Container Service (ECS) with Fargate as the serverless compute engine. The proposed architecture containerizes a Flask-based web application using Docker, automates the continuous integration and deployment (CI/CD) pipeline via AWS CodePipeline and Elastic Container Registry (ECR), and exposes a dual-portal interface-one for end-users to create and track shipments, and a secured administrative dashboard for logistics operators to update delivery statuses. Shipment state transitions trigger automated email notifications to recipients through an SMTP-based notification module. PostgreSQL is employed as the relational persistence layer, accessed through Flask-SQLAlchemy ORM. Experimental evaluations demonstrate that the proposed system achieves an average API response time of 142 milliseconds, 99.9% system availability, and supports automated horizontal scaling without manual infrastructure provisioning. The results confirm that the adoption of serverless container orchestration substantially outperforms both traditional monolithic deployments and VM-centric cloud configurations in terms of operational efficiency, deployment agility, and resource utilization.
Keywords: Serverless Computing; AWS ECS Fargate; Container Orchestration; Courier Tracking System; Flask; RESTful API; CI/CD Pipeline; Cloud-Native Architecture; Docker; Real-Time Notification.
Keywords: Serverless Computing; AWS ECS Fargate; Container Orchestration; Courier Tracking System; Flask; RESTful API; CI/CD Pipeline; Cloud-Native Architecture; Docker; Real-Time Notification.
Abstract
A Privacy-Preserving AI Companion for Student Mental Wellness: Adaptive Conversational Support with Sentiment-Aware Monitoring and Retrieval-Augmented Guidance
KONALA NAGA SOWMYA SREE, Mr.B.N. SRINIVASA GUPTA*
DOI: 10.17148/IJARCCE.2026.155310
Abstract: Psychological distress among university students has risen markedly, yet institutional counselling capacity remains limited and help-seeking is often deterred by stigma and waiting times. Conventional digital wellness applications typically rely on remote cloud services, which raises legitimate concerns about the confidentiality of highly sensitive emotional data. This paper presents a local-first, assistive software framework that delivers empathetic conversational support, continuous sentiment-aware monitoring, and document-grounded informational guidance entirely on institutionally controlled infrastructure. The platform combines a reactive single-page interface with an asynchronous service core in which all generative responses are produced by a locally hosted large language model, eliminating any transmission of student dialogue to third parties. Incoming messages are screened by a lightweight natural-language pipeline that fuses lexical polarity estimation with a curated distress-and-stress lexicon, allowing the conversational agent to adapt its tone and to escalate supportive prompts when crisis language is detected. Self-reported mood and stress check-ins have persisted and transformed into a composite well-being indicator, while an anonymised analytics module equips counsellors with cohort-level trends derived through linear regression. Curated wellness literature is made query able through a retrieval-augmented generation pipeline using sentence-transformer embeddings and a vector index. A real-time channel notifies designated staff when stress thresholds are exceeded. Evaluation shows millisecond-scale sentiment screening, sub-second time-to-first token for streamed replies, and a distress-recall of 0.93 on annotated samples. The framework offers a confidential, infrastructure-light foundation for scalable campus mental- health support and is positioned strictly as an aid rather than a clinical substitute.
Keywords: Student mental health, conversational AI, sentiment analysis, retrieval-augmented generation, large language models, privacy-preserving systems, emotion monitoring, well-being analytics
Keywords: Student mental health, conversational AI, sentiment analysis, retrieval-augmented generation, large language models, privacy-preserving systems, emotion monitoring, well-being analytics
Abstract
A Machine Learning Framework for Procrastination Detection and Academic Performance Prediction with Local Large-Language-Model Feedback
SANJANA ROKKALA, P SRINIVASA REDDY*
DOI: 10.17148/IJARCCE.2026.155311
Abstract: Academic procrastination is a pervasive self-regulation failure that undermines learning outcomes, yet it is rarely detected early enough for timely intervention. Conventional academic monitoring relies on retrospective grade analysis, which identifies struggling learners only after performance has already declined. This paper presents a machine learning framework that detects procrastination behavior and predicts academic performance from passively collected study activity, then delivers individualized guidance through a locally hosted large language model. Behavioral and temporal features-session frequency, task latency, deadline proximity, and engagement regularity are extracted from learner activity logs and used to train two complementary models: an ensemble classifier that flags procrastination risk and a gradient-boosted regressor that estimates expected performance. The backend is implemented in Python, the interactive dashboard in Node.js, and contextual feedback is generated on-device by an Ollama-served language model, preserving data privacy. On a held-out evaluation, the procrastination classifier attained 92.6% accuracy and a 0.92 F1- score, while the performance regressor achieved a coefficient of determination of 0.88. A comparative study across five algorithms confirmed that gradient boosting offered the best accuracy robustness balance. The principal contributions are an interpretable behavioral feature set for procrastination modeling, a dual classification–regression pipeline that links behavior to outcomes, and a privacy-preserving local-LLM advisory layer that converts predictions into actionable, personalized recommendations for learners and educators.
Keywords: Procrastination detection; academic performance prediction; machine learning; gradient boosting; learning analytics; large language models; Ollama; educational data mining
Keywords: Procrastination detection; academic performance prediction; machine learning; gradient boosting; learning analytics; large language models; Ollama; educational data mining