← Back to Archives
VOLUME 15, ISSUE 3, MARCH 2026
Dockerized Microservices Architecture on AWS ECS
M. Sathya Narayanan, Dr. B. Narasimhan
IMPLEMENTING INFRASTRUCTURE AS CODE WITH TARRAFORM AND DOCKER
A. Mohammed Almaas, Dr. C. Daniel Nesakumar
Automate Infrastructure using AWS CloudFormation and Docker
S SREEKA, Dr. S. SHYLAJA
Dockerize a Node.js Application and Deploy on Amazon EC2
Vighnesh Jayakumar, Dr. KS. Gowrilaksshmi
AWS Lambda for Serverless Data Aggregation and Reporting
Yuvaraj N, Mrs. Praveena
AWS CloudFormation for Automated Docker Container Deployment
Arun Prasad S, Mr. S.S. Saravana Kumar
Implement Blue/Green Deployments with Docker and AWS
Mohammed Afraz S A, Dr. S. Shylaja
Serverless REST API Todo Management System: Performance Evaluation and Cost Analysis Using AWS Lambda and DynamoDB
G. Yathishvar, Mr. S.S. Saravanakumar
SERVERLESS API WITH AWS LAMBDA AND DYNAMO DB
B. Arun, Dr. S. Thavamani
Automated Cloud Backup System Using AWS
NAVEEN PANDI, Mr. S. S. SARAVANA KUMAR
Continuous Deployment using GitLab CI/CD and Docker
R. K. Selvavishnu, Dr. B. Narasimhan
Secure Docker Containers with AWS IAM and AWS Secrets Manager
C. B. Greeshma, Mr. B. Ramesh Kumar
RishiBuild – A Scalable Chat Application Using Docker and AWS ElastiCache
A. Manikandan, Mr. S. S. Saravana Kumar
SMART SOIL IRRIGATION SYSTEM
N. Mohamed Abdul Rahman, Dr. S. Thavamani
Automating Docker Image Build and Deployment Using GitLab CI/CD
Sachinraj R, B. Ramesh Kumar
AI CyberShield Matrix: Intelligent Threat Detection and Analysis
Mr. H.M. Gaikwad, Mr. S. V. Waghmare, Varun Joshi, Mithilesh Khairnar, Shubham Kungar, Harshit Aher
Temporal Continuity in Low-Light Environments: A Ghosting-Resistant Video Enhancement Framework
K. Mithun Rithick, M. Nowshad, Dr. G. Maria Priscilla
IOT BASED TEMPERATURE CONTROL EXHAUST FAN FOR INDUSTRY
Prof. Radhika Dhengale, Abhishek Shukla, Rushikesh Sawarker, Khushi Sahu, Falguni Rokade, Komal Wagh, Tushar Shende
AI Based Study Material Generator using ChatGPT and Canva
Saran.K, Dr.K.S.Gowrilaksshmi
AUTOMATED INVOICE PROCESSING USING UIPATH AND SQLITE
V. DINESHKUMAR, Dr. S. SHYLAJA
Smart Complaint Management System for Digital Gram Panchayat
S. Sri Hariharan, Dr. N. Mahendiran
PORTABLE OPTICAL SENSOR FOR MICROPLASTIC DETECTION
Prof. Rupal Pundlik, Prachi Zode, Triveni Bisen, Sweety Gupta, Akshay Shriwastav, Nayan Yelekar
The Influencer Equation: Unraveling the Science of Social Media Impact
Vasanth P, D. Pradeep, Mrs K. Kalaivani
SignVisionAI: Real-Time Sign Language Recognition Using AI
Mr. H.M. Gaikwad, Pawar Pooja, Raundal Nisha, Sangle Anuja, Thakur Harshada
MediScan AI: A Python based Intelligent System for Chest Disease Detection and Diagnosis.
Mr. H.M. Gaikwad, Sahil Rajbanshi, Priya Vaishnav, Om Shinde, Nirjala Kachave
BUILD A REAL – TIME ANALYTICS DASHBOARD WITH AWS KINESIS AND LAMBDA
DharshiniPriya.R, Dr.S. Thavamani
AWS BATCH FOR BATCH PROCESSING OF DOCKER CONTAINER
P. Kishan, Dr . S. Thavamani
AWS LAMBDA FOR EVENT-DRIVEN MICROSERVICES ARCHITECTURE
Durga Nandhini BJ, Dr.S. Shylaja
DEPLOY A DOCKERIZED DJANGO APPLICATION ON AWS LIGHTSAIL
Kavya J, Dr.C.Daniel Nesakumar
MULTI-STAGE DOCKER BUILDS AND DEPLOYMENT ON AWS ECS
Nandhini S, Dr. B. Narasimhan
Eye Movement Driven Cursor Control Using Computer Vision Technique
Karthik. V, Dr. M.Hemalatha, M.Sc., M.Phil., Ph.D
Implement Serverless Machine Learning Inference with AWS Lambda
Abishek P, Dr. T. Pradeep
Optimization of Niche Retail Operations: Design and Implementation of the "Deera Store Management System" web application
Asif reehan J, Dr. G. Maria Priscilla
Emotion-Aware AI for Enhancing Elderly Mental Health
Abhishek R Shetty, Achith J V, Adithya Jain, Aman N, Mr. Shivaraj B G
Attention-Based Occlusion Handling for Dynamic Human-Robot Interaction in Unstructured Environments
Muthuraj S, Dr. S. Mohana
ML - Based Traffic Violation Severity Scorer
S. Ratheesh Kumar, Dr. N. Mahendiran
REAL-TIME LOG ANALYSIS WITH AWS OPENSEARCH AND DOCKER
Abinaya G, Mr. T. Pradeep
Sub-Pixel Semantic Segmentation for Precision Agriculture: Identifying Micro-Defects in Crop Foliage
Akshay K, Dr. S. Devibala Subramanian
Cybersecurity threat detection to Prevent System
Tikshanaa.P.R, Abinaya.Y, Dr.R.Nagarajan
ADAPTIVE CLOUD LEARNING TUTOR USING LLM
Prince Rosario, Shriharri, Ms A Deepika
SERVERLESS IMAGE PROCESSING WITH
AWS LAMBDA AND S, Rithika. P, Dr. S. Shylaja
Secure docker container with AWS secrets manager
M.J.Premkarthik, Dr. S. Shylaja
DOCKER SWARM CLUSTER ON AWS
A. Taniya Maria, Dr. C. Daniel Nesakumar
“AI-Based Therapist Chatbot for Emotional Support”
Santhiya.M, Ms. A. Deepika
Implementing Infrastructure Monitoring Using Docker, Prometheus and Grafana
P. Merlyn Jessica, Dr. B. Narasimhan
SECURE AWS INFRASTRUCTURE WITH DOCKER AND IAM ROLES
Sananta.M, Dr. B. Narasimhan
Automate Docker Container Management with AWS Elastic Beanstalk
Dharaneeswaran R B, Ms. A. Praveena
Serverless ELT Pipeline for Scalable Data Processing Using AWS Glue and AWS Lambda
Isacc Moses S, Mr. Daniel Nesa Kumar C
Serverless File Processing Pipeline with AWS Step Functions
Sampathkumar R S, Dr. T. Pradeep
Deploying a WordPress Site Using Docker and AWS RDS
S. Pranesh, Dr. S. Thavamani
AWS CloudTrail for Monitoring Docker Container Activity
PRATEESH S, Mr. S. S. Saravana Kumar
Serverless Data Processing Using AWS Lambda and Amazon Kinesis
H Darsan, Mr. B. Ramesh Kumar
Deploy a Dockerized E-commerce Platform on AWS ECH
Dinesh Kumar, Dr. C. Daneil Nesakumar
Serverless IoT Application Using AWS IoT Core and Lambda for Real-Time Sensor Data Processing
Dhanush K, Mr. B. Ramesh Kumar
Build a Python Flask App and Deploy Using AWS Elastic Beanstalk
Vishaal S, Mr. S.S. SaravanaKumar
Multi-Cloud Deployment using Docker and Terraform
G. Ranganayaki, Dr. B. Narasimhan
AWS Step Functions for Orchestrating Serverless Workflows
Gijith M, Praveena A
AWS LAMBDA FOR SERVERLESS MICROSERVICES ARCHITECTURE
Ponnuguru M, Mr. T. Pradeep
Smart IoT-Based Stress Detection & Autonomous Relaxation System
Prof. V Chaudhari, Devanshu A. Bidkar, Tanvi V. Bodhe, Ansh K. Kapse, Ruchi S. Rajput, Ujwal S. Sewaiwar
AI COMPLAINT-PRIORITY SYSTEM FOR LOCAL GOVERNMENT
Dr. B. Sivaranjani, Mr. B. Dharneesh, Mr. S. Dharaneesh, Mr. K. Dharshan, Mrs. K. Abirami
Deploy a Static Website on AWS S3 and Use CloudFront
Yaswanth S, Mr. B. Ramesh Kumar
NeuroSpill: Attention U-Net++ Based Oil Spill Segmentation in SAR Imagery
Aishmi Anish, Ance Koshy, Nebu Placid, Sunu Sara Antony, Neethu Thomas
The Tech-Integrated Charity Chain for Transparent Giving and Efficient Aid Delivery
Vinay S. Khapare, Pranav D. Khot, Arkaj U. Kadam, Shreyash P.Alase, Jaydeep A. Patil, Prof. Sambhaji D. Rane
DEPLOY A MULTI-REGION APPLICATION
WITH DOCKER AND AWS ROUTE, Pon Vignesh P, Gowri Lakshmi K S
School Management System
Lect. Kalyani Wankhede, Arpit Raut, Gaurav Kamble, Ashutosh Wankhede, Shubham Somkuwar, Khushi Gamey, Dimple Madavi.
SERVERLESS IMAGE RECOGNITION WITH AWS REKOGNITION AND LAMBDA
Menaka Sri S, Dr. Shylaja M
Smart Speak Home Automation
Prof. Shital Hajare, Jagdish A. Neware, Jivesh A. Karnuke, Nishant D. Mahajan, Ganesh S. Parwe, Ankita L. Barapatre
Secure Web Application Using Docker, Nginx, and AWS WAF
Jagatheeshwaran E, Dr. C. Daniel Nesakumar
CNN-Based Dorsal Hand Vein Authentication Using Triplet Loss Metric Learning
J Adithye, Abhimanyu R, Abhijith S Kurup, Adithiya S, Girija V R
AI-Based Vegetable Disease Detection
Mr. H.M. Gaikwad, Ms. V. N. Lawand, Tanuja Ahire, Shrushti Dhamode
Internet of Things (IoT): Architecture, Applications, Challenges, and Future Scope
AKHIL.V.R, SANTHOSH.V, Dr. P. RADHA
Fruit Grade
Prof. Prof. Smita Thakare, Pallavi Jankiram Solanke, Bhagyashri Mahendra Pardeshi, Tanisha Ajay Lokhande, Parth Prabhanja Naik
Insider Threat Detection using Agentic AI
R.Y. Thombare, S.V. Waghmare, Amey Malpurkar, Aditya Marathe, Chhatraraj Patil, Tanmay Gaikwad
Deepfake Audio Detection Using Hybrid Random Forest and Convolutional Neural Network Architecture
K Dharma Ratnam, P Kanaka Tulasi
An AI-Powered Interior, Exterior Designer & Layout Planner
HOD R. Y. Thombare, Mr. S. V. Waghmare, Sarthak Deshpande, Atharva Ahire, Aditya Awasarkar, Mitesh Pawar
Skin Guard Pro: AI Powered Skin Disease Detection System.
Mr. H.M. Gaikwad, Vansha Nikam, Sawant Poorva, More Lokesh, Yeola Vaishnavi
“AI ASSISTED PERSONAL ROBOT”
Prof. Bhagyashree Wankhede, Aarju S. Dhurve, Lokesh P. Kothe, Mitali S. Mankar, Pranjali A. Nagose, Swejal D. Patil
3D HUMANOID AI-SMART "AW8R”
Prof. CHANDRESH BHANGE, PRATYUSH.G. JANBHANDHU, SAMIKSHA A. GAJBHIYE, PRAKRUTI R. GAIMUKHE, TANU R. PATALE, RAVISHANKAR V. SHAHU, JAYASHRI M. MARBATE
Hybrid Cloud Deployment Using AWS EC2 and On-Premises Virtual Machine
Nagaraj L, Dr. K. S. Gowrilakshmi
CLOUD INTRUSION DETECTION SYSTEM
Mrs. K. Rajeswari, B. Thraya Gayathri, J. Hari Sravani, T. Himaja, N .P. Praharshita
Chatbot Development Using AWS Lex and Lambda
R. Sri Vignesh, Dr. B. Narasimhan
Implement Auto Scaling with Docker and AWS
EC Auto Scaling, Nishok T N, Dr. S. Thavamani
RFID And Fingerprint Based Digital Passport For Cross Border Authentication
Mr. H.M. Gaikwad, Mr. S.V. Waghmare, Saili Borse, Krutika Bhamare, Vidhi Daryani, Darshana Kothari
ROAD RULES SIMULATOR: AN INTERACTIVE LEARNING SYSTEM FOR TRAFFIC EDUCATION
Adil Niham, Goban Roshan P, Kiran Mohan, Melvin Alex, Prof. Marina Glastin
AI-Driven Second-Brain Knowledge Management System
Mrs. P. H. Nawale, Makarand Mangesh Ahire, Nikhil Nilesh Potdar, Aquino Harold Rodrigues, Shaikh Mohammed Zaid Faiyyaz
AI based underwater Remotely Operated Vehicle (ROV) for Inspection and surveillance
Mr. H.M. Gaikwad, Vaibhav Bhusare, Rida Khan, Aarya Nagpure
Adaptive Learning Technologies in Modern Education
Madhuri Ramdas Gagare, Sakshi Dhananjay Jadhav
An Automated Human Violence Detection for Surveillance using Deep Learning
Mr. H.M. Gaikwad, Sharvari Gangurde, Priyanka Moule, Sejal Patil, Aarya Patil
AgroBot: AI-Powered Smart Robotic Lawn Mower
Mr. H.M. Gaikwad, Rutuja Yogesh Deshmukh, Shubham Dnyaneshwar Kunde
Megatron: AI Powered Navigation And Information Bot
Mr. H.M. Gaikwad, Ms. V. N. Lawand, Krishi Jadhav, Abhirav Pawar, Vishal Dhule
Development of Robust Trust Management Framework for Medical IoT
K Chakravarthy Bheri, Manas Kumar Y
DriveMate AI: Personalized Intelligent Driver Assistant
Emmanuel SJ, Devnandhan M, Joel Roy, Shan Jahan, Assistant Prof. Ancy Das
Network Intrusion Detection System Using Machine Learning
Anusree P S, Aryanandha Anil R, Fathima S, Gopika K and Prof. Merlin K Thomas
PERF CLOUD: A Hybrid DTW–GRU Framework for Black-Box Virtual Machine Performance Degradation Prediction in Multi-Tenant Public Clouds
Aiswarya Lekshmi JS, Aiswarya Vijayan, Ardra Vinod, Mathews Franco, Merlin K Thomas, Ajeesh S
FarmSphere: A Smart Digital Platform for Sustainable and Connected Agriculture
Avishka Gaykar, Gauri Shelke, Bhavesh Suryavanshi, Mrs. Kalyani Kapde
Football Match Analysis With Yolo
Sachin Yadav, Vipul Chavan, Janmesh Vyas, Dr. M. M. Deshpande
MediSphere: AI-Driven Diagnostics with Precision Scoring and Therapeutic Intelligence
Dr. Jagdish.W. Bakal, Rohan Hanmant Saste, Kishor Ramesh Shirke, Abhay Rajendra Jadhav, Gaurav Ananda Lende
Timex: Timetable Automation System for Educational Resource Scheduling in Smart Campus
Prof. Vanchala Sutar, Ishita Tarak Mehta, Darshan Deepak Redekar, Aniket Ramchandra Shinde
MediScan AI: An Intelligent Vision-Based Platform for Prescription Decoding and Comprehensive Medicine Information Delivery
Mr. H.M. Gaikwad, Ms. V. N. Lawand, Vedant Dhotre, Krushna Nikam
Real-Time Expense Tracker with Smart Budget Recommendation System
Nita Ingale, Nutan Avhad, Diksha Jadhav, Pratik Pawar, Soham Gawde
Hybrid AI-Powered Legal Assistant for BNS Laws Using Structured Retrieval and Contextual Language Model Generation
Omkar Rajaram Barad, Atharava Atul Desai, Atharav Sachchidanand Bhogate, Saurabh Namdev Dhokare, Prof. (Dr.) Suresh Shirgave
AI-Driven Ferry Booking and Assistance System for Smart and Inclusive Railway Services
Viren S. Adgale, Kashish P. Jadhav, Srushti P. Jadhav, Samiksha S. Herwade, Rushikesh B. Gherade
IFOS: AN ML-POWERED REAL-TIME INTELLIGENT FILE ORGANIZATION SYSTEM
Priyanka Balaso Ingale, Mayur Narendra Gavali, Third Year B.Tech, Computer Science Engineering (Artificial Intelligence), DKTE Ichalkaranji., Final Year B.Tech, Computer Science Engineering (Artificial Intelligence), DKTE Ichalkaranji
Self-Evolving Cognitive Architectures: Bridging Neural Optimization and Consciousness Simulation for AGI
Adhithyan A, Vivek S, Febin K James, Muhammed Abnan, Asst. Prof. Jinu L
Advanced Marketplace Suggestion System For Farm Produces Using AIML
Atharv C. Chavan, Manav P. Bansode, Om L. Kamble, Rahul N. Kamble, Vaijanath M. Wale, Prof. D. D. Mahajan
Roamio: Leveraging Artificial Intelligence for Smart Traveler Matching, Itinerary Generation, and Solo Travel Safety
Vidhi Sankhe, Richa Bhardwaj, Tanishq Tamhankar, Abhay Sonkawade, Mrs. Ashwini More
Comparison Of Anthropometric Variable With Respect To Forearm Muscle Circumference Among Tribal And Non-Tribal Sportsmen With Reference To Age Groups
Gaude Pralay Rohidas
Adaptive Traffic Signal Control System Based on Future Traffic Environment
Parshwa Pravin Gebise, Aditya Shailesh Kulkarni, Devendra Santosh Gurav, Siddharth Deepak Jawale, Shivam Sanjay Dafale
AI BASED LEGAL SIMPLIFICATION AND CASE OUTCOME PREDICTION
Prof. Dnyaneshwar Thombre, Mahesh Bhosale, Vikas Maurya, Intaza Chaudhary, Mausam Yadav
AUTOMATED DATA INTEGRATION AND REAL-TIME VISUALIZATION USING
NN AND POWER BI, D VIMAL KUMAR, A. VANITHA, MISHAL ALOY J
SkillMirror: An AI-Based Communication and Interview Training Platform Using Speech and Computer Vision Analysis
Shreyas S. Gadekar, Atharv N. Mane, Prasad T. Kalavikattekar, Aniket M. Kamble, Karan P. Ghatage, Prof. Avani G Shahane
Human-Centered Machine Learning for Educational Systems: A Review of Prediction, Behavioral Analytics, and Explainable AI
Jayashree More*, Kirti Dinkar More, Sunita Nimba Deore
Design and Development of a Role-Based Industry Management System Using Modern Web Technologies.
Mohd Raed, Arhan Shaikh, Abdul Gani Kazi, Ayan Shaikh
Database-Driven Architecture for Academic Knowledge Modeling and Automated Practice Paper Generation
Haji Saad, Shaikh Mohammed Abbas, Tiwari Aditya, Naik Faiz, Shimpi Manasi
An IoT-Driven Intrusion Detection and Autonomous UAV Surveillance System with Edge-AI Verification
Blessan B Kuriakose, Aswin Prakash, Krishnajith K S, Githu Manoj, Parvathy R Nair
Explainable Fake Job Posting Detection Using OCR, Machine Learning, and AI Reasoning
RAGUNATH M, Mrs. PRADEEPHA S, Dr E. MARIAPPAN, Dr M. KALIAPPAN
DATAVISION PRO: INTERACTIVE BUSINESS INTELLIGENCE DASHBOARD
Dr. Sajja Suneel, Manipatruni Trived, Nathi Dhana Sree
Improved Customer Churn Estimation Using LSTM Networks
Dr. Sajja Suneel, Lokesh Gopal M, Harshitha P, Harshitha Reddy K
Fireguard AI: Advanced Fire Detection and Evacuation Path Guidance System Using IoT and Artificial Intelligence
Harikrishnan S, Devika MS, Alex TJ, Vishnu PS, Ananthakrishnan CS, Ann Josy John
Lok Netra: A Vision of Politics & Voting Power
Veer Panchal, Chinmay Thukarul, Shubham Toraskar, Mrs. Nisha Karolia
AI AND IOT-ENABLED AUTONOMOUS ANTI-DRONE DEFENSE SYSTEM
Aien Joel Paul, Akhil Ginoy, Albin Babu, Mahesh T.S, Ann Josy John
MULTI-LAYERED AUTHENTICATION AND ENCRYPTION FOR CLOUD SAFETY
P V Ramana Murthy, Mohammad Noman Malik, Gampa Rahul, Talari Ajay Kumar, Gurram Ravi Kiran
GraphSense-RAG: An Intelligent Multi-Modal Document Question Answering System
Dr. M. Purnachandra Rao, D. Mastan Bee, B. Navya, B. Sravani, A. Bhargavi, B. Uma
AI Career Role Recommendation System with Skill Gap Analysis, Career Progression, and Salary Simulation Using Resume Semantic Analysis
Ms. Neelam Mary Vijaya Nirmala, G. Pallavi, B. Jhansi Lakshmi, G. Ramya, J. Divya Dimple
Campus Placement Eligibility & Recruitment Outcome Prediction Using Academic and Behavioural Attributes
Mrs. M. Khamar, D. Sripriya, B. Vyshnavi, D. Gowthami, G. Sushma
Multi Model Emotion Aware Conventional Chatbot Using Facial Expression and Text Sentiment Fusion
Mrs. K. Tejaswi, Ch. Pushpa Manasa, D. Hima Sravanthi, B. Pravallika, G. Girishma
Virtual Personal Productivity Assistant with Task Automation and Intelligent Scheduling Capabilities
Mr. P. Madhubabu, Ch. Maheswari, A. Thanusha, B. Nikitha, Ch. Susmitha, Ch. Lavanya Chowdary
Conversational AI Travel Assistant for Automated Trip Planning and Destination Query Interpretation
Mrs. V. Krishna Vijaya, J. Hemanth Kumar, J. Mallikarjuna Rao, G. Veera Narayana, D. Siva Manikanta, A. Pratap
AI Driven Personalized Diet and Nutrition Recommendation System Based on Health Metrics and Food Knowledge Graphs
Mrs. J. Mounika, E. Madhan Kumar, A. Santhosh, B. Salem Raju, D. Veera Brahmam
RASA-Based End-to-End Conversational Chatbot with Intent Training and Dialogue Policy Management
Mrs. B. Kalyani, A. Datta Maheswara Das, D. Yogeswar, G. Ajay, D. Ramakrishna, G. SreeRam
Healthcare Consultation and Symptom Response Chatbot Powered by Medical NLP Dialogue Understanding
Mrs. K. Deepthi, A. Ratna Sai, CH. Karthik, G. Hemanth Kumar, G. Venkata Sai Teja
Dynamic AI Storytelling and Narrative Plot Generation Engine Using Generative Language Models
Mr. P. Madhubabu, B. Lakshmi Srinivas, B. Pavan Kumar, B. Prabhu, A. Nehemya
Real-Time Object Detection System Using YOLO-Based Vision Models
Mrs. K. Deepthi, B. Akash, Revanth Kumar, Ch. Tharun, K. Narendra
A Smart Cybersecurity Awareness Website for User Education via Courses and Gamified Learning
Rakhangi Mohammad Saif, Maniyar Mohammad, Shaikh Shahim, Shaikh Rehan, Shaikh Shahista
Smart Budget Buddy: An AI-Enhanced Personal Finance Management System Using Large Language Models and Serverless Architecture
Khan Ayaan, Shaikh Rehan, Malik Aban, Ansari Rayyan, Neha Salunkhe
Crime Report and Reward Management System Using Machine Learning
Meghana Kawale, Vrushali Nikam, Shivani Sagare, Madiha Mujawar, Tejaswini Khot, R. S. Kamble
BLOCK-CHAIN BASED DOCUMENT VERIFICATION SYSTEM USING IPFS
Prof. Swapna V. Tikore, Akash Devade, Vyankatesh Kulkarni, Sandip Pawar, Ashwin Ingle
aLLoyM: “Phase Diagram Prediction System”
Sakshi K. Kamble, Anagha G. Harshe, Soniya C. Dhupdale, Shubham U. Dharwat, Soham P. Kapileshwar
MULTI-LEVEL SIGN LANGUAGE RECOGNITION SYSTEM
NALINA P, Dr REVATHI A
Leveraging Machine Learning for Intelligent Financial Forecasting and Investment Decision Support
Mrs. Dhanashri Kulkarni, Siddhi S. Shilahar, Pushkar D. Kaslikar, Pratik Pradip Kale, Chaitanya V. Kaypure, Parth P.Kshirsagar
AI Powered Low Code/No Code Software Development
Dilip Vishwakarma, Anil Vasoya, Aruna Pawate
A Study Impact on the Influencer Marketing on Brand Trust and Purchase Intention: A Comparative Study of Millennials andGenZinIndia
Shalin Dwivedi, Ishita BhanushaliandDr. Hiren Harsora
AutoMomo–Rapid Forming Technology
Ms.G.Madhumathi, Arini.R, Jayahari.S, Nilavarashi.T
College Enquiry Chatbot UsingMachine Learning:An Intelligent Conversational System for Academic Information Retrieval
Sonu Yadav, Vivek Chavan, Sufyan Hawaldar, Samruddhi Chinchavale, Prof. Ashwini Chavan
A Multimodal AI Framework for Real-Time Audience Engagement Detection in Virtual Communication
A V Tejaswi, M Sumana Sree, S Sahithi, Dr.C.Swapna
CROP WILTING ANGLE MEASUREMENT USING OPENCV SONIA MARIA D’SOUZA1, S VINAY KUMAR REDDY2, S SAMUEL SUNDAR3, SIDDAMREDDY MOHITH REDDY4, RAVVA SAI CHARAN5
Assoc. Professor, Department of AI & ML, New Horizon CollegeofEngineering, Bangalore, India
IMPACT OF ORGANIZATIONAL POLITICS ON EMPLOYEE PERFORMANCE IN IT SECTOR IN PUNE
Abhishek Kolhe, Sayali Waghmare, Mrunali Shinde, Nandini Singharam, Pranjali Kirtane
A Comparative Study of Cardiovascular Fitness between Football Players and Non-Football Players”
Vidya Bhushan Sharma
Abstract
Dockerized Microservices Architecture on AWS ECS
M. Sathya Narayanan, Dr. B. Narasimhan
DOI: 10.17148/IJARCCE.2026.15302
Abstract: Modern software systems require flexible and scalable deployment environments to support growing workloads and complex applications. Containerization has emerged as an efficient solution that allows applications to be packaged with all required dependencies into lightweight containers. Docker is one of the most widely used containerization technologies that enables developers to build portable and consistent application environments. However, managing containers in large-scale environments requires a reliable orchestration platform.
This research presents a cloud-based deployment architecture titled “Dockerized Microservices Architecture on AWS ECS.” The system integrates Docker containerization with several Amazon Web Services (AWS) components including Amazon Elastic Container Service (ECS), Amazon Elastic Container Registry (ECR), Amazon Elastic Compute Cloud (EC2), and AWS Identity and Access Management (IAM).
Docker is used to build container images for the application, while Amazon ECR stores these images securely in private repositories. Amazon ECS is responsible for container orchestration, enabling automated deployment, monitoring, and scaling of containerized applications. Amazon EC2 provides the computing infrastructure required to run containers within the ECS cluster. IAM ensures secure authentication and role-based access control for cloud resources.
The proposed system demonstrates a scalable and secure container deployment model that improves application portability, deployment efficiency, and resource utilization. The results show that combining Docker with AWS ECS can significantly simplify cloud deployment and support modern DevOps practices for building reliable microservices-based applications.
Keywords: Docker, Microservices Architecture, AWS ECS, Cloud Deployment, Containerization, DevOps.
Abstract
IMPLEMENTING INFRASTRUCTURE AS CODE WITH TARRAFORM AND DOCKER
A. Mohammed Almaas, Dr. C. Daniel Nesakumar
DOI: 10.17148/IJARCCE.2026.15303
Abstract: The rapid growth of digital communication has increased the need for efficient methods of sharing web links across platforms. Long and complex URLs often reduce readability, are difficult to manage, and are not suitable for sharing in environments such as social media, emails, or printed documents. This paper presents SkyLink, a cloud-native URL shortening platform designed to generate concise and trackable links while providing analytics on user engagement. SkyLink is developed using a modern full-stack architecture that combines React and TypeScript for the frontend and Supabase as a Backend-as-a-Service platform. The system enables users to generate shortened URLs, track click statistics, and manage links through an interactive dashboard. Containerization using Docker ensures portability, while Infrastructure as Code implemented through Terraform allows automated provisioning of cloud resources on AWS. The system architecture emphasizes scalability, reliability, and ease of deployment. By integrating modern DevOps practices such as containerization, automated CI/CD pipelines, and serverless infrastructure, SkyLink demonstrates how contemporary web technologies can be used to build scalable and production-ready applications. Experimental evaluation shows that the system performs efficiently in generating short URLs and handling user requests with minimal latency. The proposed platform provides a cost-effective and flexible alternative to existing commercial URL shortening services.
Abstract
Automate Infrastructure using AWS CloudFormation and Docker
S SREEKA, Dr. S. SHYLAJA
DOI: 10.17148/IJARCCE.2026.15304
Abstract: Manual cloud infrastructure management leads to environment inconsistencies, deployment errors, and significant operational overhead. Existing productivity timer applications suffer from poor visual engagement and lack modern deployment practices. This paper presents Focus Timer, a React 19 web application with a glassmorphic interface, adaptive time-of-day theming, and SVG-based circular progress visualization, deployed to AWS ECS Fargate through a fully automated pipeline using Docker and AWS CloudFormation.
The multi-stage Docker build produces a production image of approximately 25 MB. The Vite 7 build generates a gzipped bundle under 50 KB with sub-second page load. CloudFormation templates reproducibly provision an ECS Cluster, IAM Execution Role, and Fargate Task Definition across any AWS region. Automated PowerShell scripts complete the full pipeline from source to running cloud service in under six minutes, eliminating all manual console steps. All 20 functional test cases passed across Chrome, Firefox, Edge, and Safari.
Keywords: React 19, Vite 7, AWS CloudFormation, Docker, ECS Fargate, Infrastructure-as-Code, Glassmorphism, SVG Animations, ECR.
Abstract
Dockerize a Node.js Application and Deploy on Amazon EC2
Vighnesh Jayakumar, Dr. KS. Gowrilaksshmi
DOI: 10.17148/IJARCCE.2026.15305
Abstract: Cloud computing and containerization technologies have fundamentally transformed modern software application deployment. Traditional deployment methods required manual server configuration, dependency management, and environment maintenance—processes that were time-consuming and error-prone. This project demonstrates the complete, end-to-end process of Dockerizing a Node.js application and deploying it on an Amazon Elastic Compute Cloud (EC2) instance. Docker encapsulates the application and its runtime environment into a portable container image, ensuring consistent behavior across environments. The project covers EC2 instance provisioning on Amazon Linux 2023, Docker Engine installation, container lifecycle management, AWS Security Group configuration, port mapping, and comprehensive deployment verification through terminal commands and browser-based testing. Results confirm successful deployment with the application accessible via public IPv4, validating the effectiveness of containerized cloud deployment as a modern DevOps solution.
Keywords: Docker, Node.js, Amazon Web Services (AWS), Amazon EC2, Containerization, Cloud Deployment, DevOps, Docker Image, Dockerfile, Security Groups, Port Mapping, Amazon Linux 2023, CI/CD.
Abstract
AWS Lambda for Serverless Data Aggregation and Reporting
Yuvaraj N, Mrs. Praveena
DOI: 10.17148/IJARCCE.2026.15306
Abstract: In the modern digital world, organizations generate large volumes of data every day. Managing and analyzing this data efficiently is a critical challenge for businesses. Traditional data processing systems rely on manual effort and server based infrastructure, making them slow, expensive, and difficult to scale. This project proposes a Serverless Data Aggregation and Automated Reporting System using Amazon Web Services (AWS). The system leverages AWS Lambda for serverless data processing, Amazon DynamoDB for scalable cloud based data storage, Amazon S3 for secure report storage, Amazon EventBridge for automated scheduling, and Amazon CloudWatch for real time monitoring. Sales transaction data stored in DynamoDB is automatically processed by Lambda functions that generate structured CSV reports uploaded to S3 at scheduled intervals. The proposed system eliminates manual reporting overhead, reduces infrastructure costs, and provides a scalable, reliable, and cost effective solution for real world data aggregation and business reporting requirements.
Keywords: AWS Lambda, Serverless Computing, Amazon DynamoDB, Amazon S3, Cloud Computing, Data Aggregation, Automated Reporting, Amazon EventBridge, Python.
Abstract
AWS CloudFormation for Automated Docker Container Deployment
Arun Prasad S, Mr. S.S. Saravana Kumar
DOI: 10.17148/IJARCCE.2026.15307
Abstract: In modern cloud computing environments, automation and scalability are essential for application deployment. Traditional deployment methods involve manual server configuration, dependency installation, and repeated environment setup, which lead to configuration errors and increased deployment time. This paper presents a system titled 'AWS CloudFormation for Automated Docker Container Deployment,' which automates the process of deploying containerized applications using cloud infrastructure. A Python Flask web application is containerized using Docker and stored in Amazon Elastic Container Registry (ECR). Deployment is managed through Amazon Elastic Container Service (ECS), while AWS CloudFormation automates infrastructure provisioning following the Infrastructure as Code (IaC) paradigm. A CI/CD pipeline using AWS CodePipeline and AWS CodeBuild enables automatic build and deployment on code changes. An Application Load Balancer (ALB) ensures high availability by distributing user traffic across running containers. Experimental results demonstrate that the proposed system significantly reduces deployment time and eliminates manual configuration errors.
Keywords: AWS CloudFormation, Docker, Amazon ECS, Amazon ECR, CI/CD Pipeline, Infrastructure as Code, Container Deployment, Application Load Balancer, DevOps, Cloud Computing.
Abstract
Implement Blue/Green Deployments with Docker and AWS
Mohammed Afraz S A, Dr. S. Shylaja
DOI: 10.17148/IJARCCE.2026.15308
Abstract: Modern software systems demand deployment strategies that eliminate downtime and ensure continuous service availability. Traditional in-place deployments cause service interruptions ranging from minutes to hours, directly impacting user experience, customer trust, and business revenue. This project implements a Blue/Green Deployment strategy for a containerized Node.js e-commerce web application hosted on Amazon Web Services (AWS). The system leverages Docker for containerization, AWS Elastic Container Registry (ECR) for versioned image storage, AWS EC2 for compute infrastructure, and an AWS Application Load Balancer (ALB) for intelligent traffic routing between Blue and Green environments. The Blue environment serves live production traffic, while the Green environment hosts the updated application version. Following automated health-check validation, the ALB listener rule is updated to switch traffic from Blue to Green in under one second, achieving true zero-downtime deployment. Emergency rollback reverses the switch in under 30 seconds without service interruption. Evaluation results confirm that the system sustains over 450 requests per second with average response times below 115 ms, while all health check test cases pass successfully. The project provides a reproducible, enterprise-grade reference architecture for DevOps practitioners adopting continuous deployment pipelines.
Keywords: Blue/Green Deployment, Docker, AWS EC2, AWS ECR, Application Load Balancer, Zero-Downtime Deployment, Node.js, DevOps, Containerization, Cloud Computing, Continuous Delivery.
Abstract
Serverless REST API Todo Management System: Performance Evaluation and Cost Analysis Using AWS Lambda and DynamoDB
G. Yathishvar, Mr. S.S. Saravanakumar
DOI: 10.17148/IJARCCE.2026.15309
Abstract: This paper presents an empirical evaluation of serverless computing architecture through the development and deployment of a REST API-based todo management system using AWS cloud services. The system leverages AWS Lambda for serverless compute, Amazon API Gateway for HTTP endpoint management, and Amazon DynamoDB for NoSQL data persistence. Through systematic performance testing and cost analysis conducted over a 30-day operational period, we demonstrate the practical advantages and limitations of serverless architecture for web applications. Performance measurements show average response latencies of 65-78 milliseconds for warm Lambda executions across CRUD operations, with success rates exceeding 99.7%. Cost analysis reveals monthly operational expenses of approximately $1.33 for 50,000 requests, representing a 97% reduction compared to equivalent EC2-based infrastructure. The study addresses challenges including cold start latency mitigation, CORS configuration, and DynamoDB schema design for NoSQL environments. Implementation artifacts include five Lambda functions totaling 850 lines of Python code and a responsive web frontend built with vanilla JavaScript. This work contributes practical insights for developers adopting serverless technologies and validates theoretical serverless computing advantages through measurable real-world deployment metrics.
Keywords: Serverless Computing, AWS Lambda, REST API, Amazon DynamoDB, Performance Evaluation, Cost Analysis, Function-as-a-Service, Cloud-Native Architecture.
Abstract
SERVERLESS API WITH AWS LAMBDA AND DYNAMO DB
B. Arun, Dr. S. Thavamani
DOI: 10.17148/IJARCCE.2026.15310
Abstract: Modern web applications require scalable and efficient backend systems to manage large volumes of requests and data. Traditional server-based architectures often require complex infrastructure management and maintenance. To address these limitations, serverless computing has emerged as an effective solution for building scalable and cost-efficient APIs.
This paper presents the design and implementation of a Serverless API using AWS Lambda and DynamoDB. The system leverages Amazon Web Services to build a fully serverless architecture that eliminates the need for server management while providing automatic scaling and high availability. AWS Lambda is used to execute backend logic, while Amazon DynamoDB serves as a highly scalable NoSQL database for data storage.
The proposed system enables developers to build lightweight, scalable APIs that can process requests efficiently without maintaining traditional server infrastructure. The evaluation results demonstrate that serverless architecture reduces operational overhead, improves scalability, and optimizes resource utilization.
Keywords: Serverless Computing, AWS Lambda, DynamoDB, Cloud Computing, REST API, Serverless Architecture
Abstract
Automated Cloud Backup System Using AWS
NAVEEN PANDI, Mr. S. S. SARAVANA KUMAR
DOI: 10.17148/IJARCCE.2026.15311
Abstract: With the increasing dependence on digital data, the need for reliable backup and data recovery mechanisms has become critical. Data loss may occur due to hardware failure, accidental deletion, cyber threats, or system crashes. Traditional backup systems often rely on manual copying or scheduled backups, which are time-consuming and prone to human errors.
This research proposes an Automated Cloud Backup System using Amazon Web Services (AWS) that ensures secure and automatic replication of files in a cloud environment. The system utilizes Amazon S3 for scalable object storage, AWS Lambda for serverless automation, IAM for secure access management, and CloudWatch for monitoring system activity.
Whenever a file is uploaded to the primary S3 bucket, an event notification automatically triggers the Lambda function. The Lambda function then replicates the file into a secondary backup bucket, ensuring that a duplicate copy of the data is always available.
This event-driven architecture eliminates manual backup processes and improves data reliability.
The experimental results demonstrate that the proposed system successfully performs automated file replication with minimal latency and high reliability. The system ensures efficient data protection and provides a scalable backup solution for modern cloud-based applications.
Keywords: Cloud Computing, Data Backup, Amazon Web Services, AWS Lambda, Amazon S3, Serverless Architecture
Abstract
Continuous Deployment using GitLab CI/CD and Docker
R. K. Selvavishnu, Dr. B. Narasimhan
DOI: 10.17148/IJARCCE.2026.15312
Abstract: Continuous Deployment (CD) is a modern DevOps practice that enables automatic building, testing, and deployment of applications whenever changes are pushed to the source code repository. This project focuses on implementing Continuous Deployment using GitLab CI/CD pipelines and Docker container technology to automate the software delivery process. The system integrates version control, automated pipeline execution, container image creation, testing, and deployment into a seamless workflow. When a developer pushes code to the main branch, the GitLab CI/CD pipeline is automatically triggered. The application is built into a Docker image, tested, stored in a container registry, and deployed to a target server without manual intervention. By containerizing the application using Docker, consistency across development, testing, and production environments is ensured. GitLab CI/CD manages the automation stages including build, test, and deploy, reducing human error and improving deployment speed and reliability. The implementation demonstrates how automated pipelines improve software quality, enhance team productivity, and support faster release cycles, providing a scalable foundation for future enhancements such as Kubernetes-based deployment and automated monitoring systems.
Keywords: Continuous Deployment, GitLab CI/CD, Docker, DevOps, Containerization, Automation, Pipeline, Software Delivery.
Abstract
Secure Docker Containers with AWS IAM and AWS Secrets Manager
C. B. Greeshma, Mr. B. Ramesh Kumar
DOI: 10.17148/IJARCCE.2026.15313
Abstract: Containerization has become a fundamental approach in modern cloud computing, offering portability, scalability, and operational efficiency for application deployment. However, securing containerized applications remains a critical challenge, particularly when managing sensitive data such as API keys, database credentials, and authentication tokens. This project focuses on implementing robust security practices for Docker containers deployed in cloud environments by integrating AWS Identity and Access Management (IAM) and AWS Secrets Manager. IAM is used to enforce fine-grained access control through roles and policies, ensuring that containers operate under the principle of least privilege. AWS Secrets Manager provides a secure mechanism for storing, managing, and dynamically retrieving sensitive credentials without hardcoding them into application code or Docker images. The implementation demonstrates a secure and scalable architecture that eliminates credential exposure risks, enhances compliance with cloud security standards, and improves overall system reliability. Results confirm that the system successfully achieves zero hardcoded credentials, encrypted secret storage, role-based access control, and automatic secret rotation capability, providing a reproducible reference architecture for securing containerized cloud-native applications.
Keywords: Docker, AWS IAM, AWS Secrets Manager, Container Security, DevOps, Cloud Security, Least Privilege, Credential Management.
Keywords: Docker, AWS IAM, AWS Secrets Manager, Container Security, DevOps, Cloud Security, Least Privilege, Credential Management.
Abstract
RishiBuild – A Scalable Chat Application Using Docker and AWS ElastiCache
A. Manikandan, Mr. S. S. Saravana Kumar
DOI: 10.17148/IJARCCE.2026.15314
Abstract: Real-time communication has become a fundamental requirement in modern cloud-based applications. This project presents RishiBuild, a scalable chat application designed and implemented using WebSocket-based communication, Docker containerization, and Amazon Web Services (AWS) cloud infrastructure. The backend server, built with Node.js and Express.js, is deployed inside a Docker container hosted on an AWS EC2 instance. AWS ElastiCache (Redis) is integrated as a distributed in-memory caching layer, enabling fast message synchronization across multiple server instances through a Publish/Subscribe (Pub/Sub) mechanism. The system supports horizontal scalability, environment consistency, and secure cloud deployment through controlled AWS Security Group configuration. Testing results confirm that the application successfully handles concurrent WebSocket connections with low latency and stable performance. The project demonstrates practical implementation of modern DevOps practices, distributed caching, and cloud-native architecture suitable for real-world deployment scenarios.
Keywords: WebSocket, Docker, AWS ElastiCache, Redis, Node.js, Real-Time Communication, Cloud Deployment, Distributed Caching, Scalability.
Keywords: WebSocket, Docker, AWS ElastiCache, Redis, Node.js, Real-Time Communication, Cloud Deployment, Distributed Caching, Scalability.
Abstract
SMART SOIL IRRIGATION SYSTEM
N. Mohamed Abdul Rahman, Dr. S. Thavamani
DOI: 10.17148/IJARCCE.2026.15315
Abstract: Agriculture plays a vital role in supporting human life and economic development. Efficient irrigation is essential for maintaining healthy crop growth and conserving water resources. Traditional irrigation methods often depend on manual observation, which can lead to over-irrigation or under-irrigation. This paper presents the design and implementation of a Smart Soil Irrigation System that automates irrigation based on soil moisture levels. The system uses a soil moisture sensor to continuously monitor soil conditions and a microcontroller to control a water pump through a relay module. When the soil becomes dry, the system automatically activates irrigation, and when the desired moisture level is reached, the pump stops. The proposed system operates locally without relying on internet connectivity or cloud infrastructure, making it cost-effective and suitable for small-scale farmers. Experimental evaluation shows that the system effectively maintains appropriate soil moisture levels and reduces unnecessary water usage.
Keywords: Smart Irrigation, Soil Moisture Sensor, Automated Irrigation System, Microcontroller, Precision Agriculture
Keywords: Smart Irrigation, Soil Moisture Sensor, Automated Irrigation System, Microcontroller, Precision Agriculture
Abstract
Automating Docker Image Build and Deployment Using GitLab CI/CD
Sachinraj R, B. Ramesh Kumar
DOI: 10.17148/IJARCCE.2026.15316
Abstract: The exponential growth of cloud-native applications and microservices architectures demands robust, scalable, and automated deployment mechanisms. This paper presents a comprehensive CI/CD-based automated pipeline system for Docker image building and application deployment using GitLab pipelines. The proposed system integrates containerization using Docker with automated workflows defined in GitLab's YAML-based configuration system. By eliminating manual intervention in the build, test, and deployment lifecycle, the system significantly reduces human error, ensures consistency in runtime environments, and accelerates software delivery velocity. Key aspects include Dockerfile authoring strategies, GitLab Runner configuration, registry management, and environment-specific deployment triggers. Experimental results demonstrate that pipeline execution time, deployment reliability, and developer productivity are substantially improved compared to conventional manual deployments. This work contributes a replicable architecture that aligns with modern DevOps practices, supporting continuous integration, continuous delivery, and infrastructure-as- code paradigms.
Keywords: Docker, GitLab CI/CD, DevOps, Containerization, Automated Deployment, Microservices, Infrastructure as Code, GitLab Runner, Continuous Integration, Continuous Delivery
Keywords: Docker, GitLab CI/CD, DevOps, Containerization, Automated Deployment, Microservices, Infrastructure as Code, GitLab Runner, Continuous Integration, Continuous Delivery
Abstract
AI CyberShield Matrix: Intelligent Threat Detection and Analysis
Mr. H.M. Gaikwad, Mr. S. V. Waghmare, Varun Joshi, Mithilesh Khairnar, Shubham Kungar, Harshit Aher
DOI: 10.17148/IJARCCE.2026.15317
Abstract: In the modern digital era, the cybersecurity threat landscape has become increasingly volatile, with sophisticated attacks such as zero-day phishing, AI-generated deepfakes, and malware bypassing traditional signature- based defenses. Comprehensive security solutions, such as Security Operations Centers (SOCs), remain prohibitively expensive and complex for individuals and small-to-medium enterprises (SMEs). To bridge this gap, this paper presents the "AI CyberShield Matrix," a unified, web-based cybersecurity toolkit powered by a Hybrid Artificial Intelligence architecture. The system integrates 14 specialized security modules into a single, user-friendly dashboard. By synergizing Supervised Learning (Random Forest) for phishing detection, Deep Learning (Convolutional Neural Networks) for deepfake analysis, and Unsupervised Learning (Isolation Forest) for User Entity Behavior Analytics (UEBA), the system provides a robust "Defense-in-Depth" mechanism. Experimental results demonstrate that this consolidated approach effectively democratizes advanced threat detection, offering real-time, highly accurate forensic analysis and remediation strategies for non-expert users.
Keywords: Cybersecurity, Hybrid Artificial Intelligence, Phishing Detection, Deepfake Analysis, UEBA, Convolutional Neural Networks (CNN), Machine Learning.
Keywords: Cybersecurity, Hybrid Artificial Intelligence, Phishing Detection, Deepfake Analysis, UEBA, Convolutional Neural Networks (CNN), Machine Learning.
Abstract
Temporal Continuity in Low-Light Environments: A Ghosting-Resistant Video Enhancement Framework
K. Mithun Rithick, M. Nowshad, Dr. G. Maria Priscilla
DOI: 10.17148/IJARCCE.2026.15318
Abstract: Low-light video enhancement has made considerable progress through deep learning, yet a persistent and often underappreciated failure mode remains: the introduction of ghosting artifacts across successive frames when scene content or camera motion causes temporal misalignment. This paper presents a Ghosting-Resistant Video Enhancement Framework (GR-VEF) that addresses temporal discontinuity as a first-class concern rather than a post-hoc correction. The proposed architecture couples an illumination-adaptive frequency decomposition module with a motion-aware temporal fusion network, coordinated through what we term a Coherence Gating mechanism. Unlike frame-by-frame enhancement pipelines, GR-VEF explicitly models inter-frame dependencies at multiple temporal scales, penalising enhancement choices that introduce perceptible flickering or double-edge artefacts even when individual frame quality metrics improve. On synthetic low-light sequences derived from the LOL-Video and SMID datasets, and on a purposebuilt evaluation corpus of real surveillance footage captured at 0.1–3 lux, GR-VEF achieves a PSNR improvement of 2.1–3.4 dB over the nearest competing method while reducing the Ghosting Artifact Index (GAI) by 38–52 percent. Qualitative inspection confirms substantially smoother temporal transitions, particularly in scenes with fast-moving foreground objects, which historically represent the hardest case for enhancement methods that rely on naively aligned reference frames.
Keywords: low-light video enhancement, ghosting artefacts, temporal coherence, deep learning, video processing, illumination normalisation, motion-aware fusion.
Keywords: low-light video enhancement, ghosting artefacts, temporal coherence, deep learning, video processing, illumination normalisation, motion-aware fusion.
Abstract
IOT BASED TEMPERATURE CONTROL EXHAUST FAN FOR INDUSTRY
Prof. Radhika Dhengale, Abhishek Shukla, Rushikesh Sawarker, Khushi Sahu, Falguni Rokade, Komal Wagh, Tushar Shende
DOI: 10.17148/IJARCCE.2026.15319
Abstract: The rapid advancement of industrial automation necessitates robust environmental monitoring systems to ensure operational safety and equipment longevity. Conventional exhaust systems often rely on manual intervention or simple thermostat-based triggers, which lack precision and fail to address hazardous gas accumulation or remote accessibility. To mitigate these challenges, this paper proposes an intelligent, IoT-based environmental control system designed for real-time monitoring and automated thermal management. The proposed solution integrates an ESP32 microcontroller as the central processing unit, leveraging its dual-core capabilities and built-in WiFi for seamless cloud connectivity via the MQTT protocol on the HiveMQ platform. The system incorporates a multi-sensor array comprising a DHT11 for ambient temperature and humidity tracking, a highprecision DS18B20 sensor for localized exhaust thermal monitoring, and an MQ-5 sensor for detection of LPG and combustible gas leakages.
A dual-mode control logic is implemented, allowing the system to operate autonomously based on pre-defined safety thresholds—specifically triggering ventilation when ambient temperatures exceed 30°C or gas levels surpass 500 PPM— while simultaneously providing a manual override feature through a professional-grade web dashboard. This dashboard facilitates real-time data visualization through dynamic charts and allows for the remote adjustment of exhaust thermal limits. Expected results indicate a significant reduction in response time to critical environmental changes and enhanced preventive maintenance through detailed PDF data logging. The practical benefits of this system include improved industrial safety, energy efficiency through demand-based fan operation, and the ability to monitor hazardous environments from any location, making it a scalable solution for modern smart infrastructures. factories and laboratory
A dual-mode control logic is implemented, allowing the system to operate autonomously based on pre-defined safety thresholds—specifically triggering ventilation when ambient temperatures exceed 30°C or gas levels surpass 500 PPM— while simultaneously providing a manual override feature through a professional-grade web dashboard. This dashboard facilitates real-time data visualization through dynamic charts and allows for the remote adjustment of exhaust thermal limits. Expected results indicate a significant reduction in response time to critical environmental changes and enhanced preventive maintenance through detailed PDF data logging. The practical benefits of this system include improved industrial safety, energy efficiency through demand-based fan operation, and the ability to monitor hazardous environments from any location, making it a scalable solution for modern smart infrastructures. factories and laboratory
Abstract
AI Based Study Material Generator using ChatGPT and Canva
Saran.K, Dr.K.S.Gowrilaksshmi
DOI: 10.17148/IJARCCE.2026.15320
Abstract: :The rapid growth of digital education has created a demand for tools that can help students generate and organize study materials efficiently. Traditional methods of preparing study materials often require significant time and effort, as students must manually search for information, summarize content, and create revision materials. To address this challenge, the AI Based Study Material Generator using ChatGPT and Canva has been developed as an intelligent system that automatically generates structured study materials from user-provided topics.
The system utilizes artificial intelligence models capable of natural language processing to analyze user input and produce educational content such as summaries, flashcards, quizzes, and visual diagrams. By integrating ChatGPT for text generation and Canva for visual design, the system provides a comprehensive learning tool that supports both textual and visual learning styles.
Users can simply enter a topic through a web interface, and the system automatically processes the request and generates relevant study materials. The generated materials help students understand complex concepts quickly and support effective revision.
The proposed system improves learning efficiency by reducing the time required for manual note preparation and providing structured educational resources. It demonstrates the potential of artificial intelligence technologies in enhancing modern education and supporting students in their academic learning process.
Keywords: Artificial Intelligence, Study Material Generation, ChatGPT, Educational Technology, Natural Language Processing, Automated Learning System, Flashcards, Quiz Generation, Digital Learning, Canva Integration.
The system utilizes artificial intelligence models capable of natural language processing to analyze user input and produce educational content such as summaries, flashcards, quizzes, and visual diagrams. By integrating ChatGPT for text generation and Canva for visual design, the system provides a comprehensive learning tool that supports both textual and visual learning styles.
Users can simply enter a topic through a web interface, and the system automatically processes the request and generates relevant study materials. The generated materials help students understand complex concepts quickly and support effective revision.
The proposed system improves learning efficiency by reducing the time required for manual note preparation and providing structured educational resources. It demonstrates the potential of artificial intelligence technologies in enhancing modern education and supporting students in their academic learning process.
Keywords: Artificial Intelligence, Study Material Generation, ChatGPT, Educational Technology, Natural Language Processing, Automated Learning System, Flashcards, Quiz Generation, Digital Learning, Canva Integration.
Abstract
AUTOMATED INVOICE PROCESSING USING UIPATH AND SQLITE
V. DINESHKUMAR, Dr. S. SHYLAJA
DOI: 10.17148/IJARCCE.2026.15321
Abstract: The rapid growth of digital business operations has led to a significant increase in the volume of invoices that organizations must process daily. Traditional invoice handling methods rely heavily on manual data entry, which is time- consuming, prone to human errors, and inefficient when dealing with large numbers of documents. This project presents an Automated Invoice Processing System developed using UiPath Robotic Process Automation (RPA) and SQLite database technology. The system automatically processes invoice documents in PDF format by extracting relevant information using Optical Character Recognition (OCR). Key invoice details such as invoice number, invoice date, total amount, vendor type, and status are identified and validated through rule-based logic. The extracted data is then stored in a structured SQLite database for efficient management and retrieval. The system also includes mechanisms for duplicate detection, error handling, and automated file segregation. Valid invoices are stored in the processed folder, while invoices with missing or incorrect information are moved to an error folder for review. By automating repetitive tasks such as document reading, data extraction, validation, and storage, the proposed system significantly improves operational efficiency and accuracy. The implementation demonstrates how RPA combined with OCR technology can transform manual document processing into a reliable and scalable automated workflow, making it suitable for small and medium-scalebusiness-environments.
Keywords: Robotic Process Automation (RPA), UiPath, Invoice Processing, Optical Character Recognition (OCR), SQLite Database, Document Automation, Data Extraction, Workflow Automation
Keywords: Robotic Process Automation (RPA), UiPath, Invoice Processing, Optical Character Recognition (OCR), SQLite Database, Document Automation, Data Extraction, Workflow Automation
Abstract
Smart Complaint Management System for Digital Gram Panchayat
S. Sri Hariharan, Dr. N. Mahendiran
DOI: 10.17148/IJARCCE.2026.15322
Abstract: Digital governance has a vital role to play in enhancing public services in rural areas. In the context of Gram Panchayat, there are certain difficulties in efficiently handling citizen complaints due to manual processing, lack of transparency, and delayed response systems. This paper introduces a Smart Complaint Management System for Digital Gram Panchayat, which helps citizens to file complaints online while ensuring efficient handling of complaints by the administrators. The suggested system utilizes web technologies to efficiently process complaints, track, and notify the citizens regarding the status of the complaint, helping to enhance transparency, accountability, and efficiency in rural governance systems.
Keywords: Digital Governance, Gram Panchayat, Complaint Management System, E-Governance, Web Application [10] Government of India, "Digital India Programme and Rural Development," Ministry of Electronics and Information Technology, New Delhi, India, 2020.
Keywords: Digital Governance, Gram Panchayat, Complaint Management System, E-Governance, Web Application [10] Government of India, "Digital India Programme and Rural Development," Ministry of Electronics and Information Technology, New Delhi, India, 2020.
Abstract
PORTABLE OPTICAL SENSOR FOR MICROPLASTIC DETECTION
Prof. Rupal Pundlik, Prachi Zode, Triveni Bisen, Sweety Gupta, Akshay Shriwastav, Nayan Yelekar
DOI: 10.17148/IJARCCE.2026.15323
Abstract: Microplastic contamination in freshwater systems has emerged as a global ecological crisis, necessitating rapid and decentralized monitoring solutions. Traditional laboratory methods, such as FTIR and Raman spectroscopy, provide high accuracy but are limited by high costs, extensive sample preparation, and lack of portability. This paper presents the design and implementation of a portable, IoT-integrated optical sensor for real-time microplastic detection. The system utilizes an ESP32 microcontroller.
Abstract
The Influencer Equation: Unraveling the Science of Social Media Impact
Vasanth P, D. Pradeep, Mrs K. Kalaivani
DOI: 10.17148/IJARCCE.2026.15324
Abstract: - In the digital world that we are in today, influencer marketing is an essential part of the process to improve the growth and sales of the brand. The main focus of this project is to determine the influence of the different influencers on the growth of the brand using data analysis. The data that is provided contains information about the campaign ID, platform, influencer type, campaign type, start date, stop date, engagement, estimated reach, product sales, and campaign duration. The main focus of the project is to determine the influence of the engagement of the influencers and the type of influencers on the sales of the product using statistical and machine learning methods. However, the main focus is to determine the percentage value of the influence of the individual influencers on the sales growth.
Keywords: Influencer Marketing, Brand Growth, Data Analytics, Random Forest Regression, Campaign
Keywords: Influencer Marketing, Brand Growth, Data Analytics, Random Forest Regression, Campaign
Abstract
SignVisionAI: Real-Time Sign Language Recognition Using AI
Mr. H.M. Gaikwad, Pawar Pooja, Raundal Nisha, Sangle Anuja, Thakur Harshada
DOI: 10.17148/IJARCCE.2026.15325
Abstract: Communication is a fundamental necessity for human interaction. However, individuals with hearing and speech impairments often face significant difficulties in expressing themselves and understanding others due to the lack of common communication platforms. Sign language serves as the primary medium of communication for such individuals, but the majority of the population is not trained to understand it. This creates a communication gap, limiting social interaction, education, employment opportunities, and access to public services.
The project “Real-Time Sign Language Recognition Using Artificial Intelligence” aims to bridge this communication gap by developing an intelligent system capable of recognizing hand gestures in real time and converting them into meaningful text or speech output. The proposed system utilizes advanced Artificial Intelligence techniques, particularly Deep Learning models such as Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN), combined with Computer Vision methods for gesture detection and recognition.
The system captures live video input through a webcam, processes the frames using OpenCV and MediaPipe for hand detection and tracking, extracts relevant features, and classifies gestures using trained AI models. The recognized gestures are then translated into text and optionally into speech, enabling seamless and natural communication. The modular design of the system ensures high accuracy, scalability, and flexibility under varying environmental conditions such as lighting and background complexity.
This project contributes significantly to assistive technology by promoting inclusivity and accessibility. It finds applications in education, healthcare, public services, workplaces, and personal communication. By integrating AI, deep learning, and real-time processing, the system provides a cost-effective, efficient, and user-friendly solution that empowers hearing- impaired individuals to communicate independently and confidently in everyday life.
Keywords: Artificial Intelligence, Deep Learning, Convolutional Neural Networks, Computer Vision, MediaPipe, OpenCV, Real-Time Gesture Recognition, Sign Language, Assistive Technology.
The project “Real-Time Sign Language Recognition Using Artificial Intelligence” aims to bridge this communication gap by developing an intelligent system capable of recognizing hand gestures in real time and converting them into meaningful text or speech output. The proposed system utilizes advanced Artificial Intelligence techniques, particularly Deep Learning models such as Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN), combined with Computer Vision methods for gesture detection and recognition.
The system captures live video input through a webcam, processes the frames using OpenCV and MediaPipe for hand detection and tracking, extracts relevant features, and classifies gestures using trained AI models. The recognized gestures are then translated into text and optionally into speech, enabling seamless and natural communication. The modular design of the system ensures high accuracy, scalability, and flexibility under varying environmental conditions such as lighting and background complexity.
This project contributes significantly to assistive technology by promoting inclusivity and accessibility. It finds applications in education, healthcare, public services, workplaces, and personal communication. By integrating AI, deep learning, and real-time processing, the system provides a cost-effective, efficient, and user-friendly solution that empowers hearing- impaired individuals to communicate independently and confidently in everyday life.
Keywords: Artificial Intelligence, Deep Learning, Convolutional Neural Networks, Computer Vision, MediaPipe, OpenCV, Real-Time Gesture Recognition, Sign Language, Assistive Technology.
Abstract
MediScan AI: A Python based Intelligent System for Chest Disease Detection and Diagnosis.
Mr. H.M. Gaikwad, Sahil Rajbanshi, Priya Vaishnav, Om Shinde, Nirjala Kachave
DOI: 10.17148/IJARCCE.2026.15326
Abstract: MediScan AI is an advanced AI-powered chest disease detection and clinical decision support system designed to enhance diagnostic accuracy and optimize healthcare workflows. The system integrates the EfficientNetV2B0 deep learning architecture with Explainable AI (Grad-CAM) and Google Gemini AI to provide accurate multi-class chest disease classification and automated clinical recommendations.
The proposed system detects over 50 chest pathologies—including Pneumonia, Tuberculosis, Lung Cancer, and COVID-19—from chest X-ray images with a classification accuracy of 99.2%. Grad-CAM heatmap visualization improves model transparency by highlighting disease-affected regions, enabling radiologists to validate AI decisions.
The architecture consists of a Flask-based backend for model deployment and a React 19 frontend for real-time interaction and patient management. Additionally, Google Gemini AI enhances the system by generating structured treatment suggestions and preliminary prescriptions, transforming the platform into a Clinical Decision Support System (CDSS).
MediScan AI bridges the gap between medical imaging and intelligent automation, reducing diagnostic errors, improving workflow efficiency, and supporting healthcare professionals with explainable and reliable AI assistance.
Keywords: Chest Disease Detection, EfficientNetV2, Deep Learning, Medical Imaging, Grad-CAM, Explainable AI (XAI), Clinical Decision Support System (CDSS), Flask, ReactJS, Google Gemini AI
The proposed system detects over 50 chest pathologies—including Pneumonia, Tuberculosis, Lung Cancer, and COVID-19—from chest X-ray images with a classification accuracy of 99.2%. Grad-CAM heatmap visualization improves model transparency by highlighting disease-affected regions, enabling radiologists to validate AI decisions.
The architecture consists of a Flask-based backend for model deployment and a React 19 frontend for real-time interaction and patient management. Additionally, Google Gemini AI enhances the system by generating structured treatment suggestions and preliminary prescriptions, transforming the platform into a Clinical Decision Support System (CDSS).
MediScan AI bridges the gap between medical imaging and intelligent automation, reducing diagnostic errors, improving workflow efficiency, and supporting healthcare professionals with explainable and reliable AI assistance.
Keywords: Chest Disease Detection, EfficientNetV2, Deep Learning, Medical Imaging, Grad-CAM, Explainable AI (XAI), Clinical Decision Support System (CDSS), Flask, ReactJS, Google Gemini AI
Abstract
BUILD A REAL – TIME ANALYTICS DASHBOARD WITH AWS KINESIS AND LAMBDA
DharshiniPriya.R, Dr.S. Thavamani
DOI: 10.17148/IJARCCE.2026.15327
Abstract: Real-time weather monitoring plays an important role in many sectors including agriculture, transportation, disaster management, and environmental monitoring. Traditional weather systems generally rely on periodic updates which may not provide immediate information when environmental conditions change rapidly. With the advancement of cloud computing and streaming technologies, modern applications are capable of processing and visualizing data continuously as events occur. This paper presents WeathX, a Real-Time Weather Analytics Dashboard that demonstrates a cloud-inspired streaming architecture for continuous weather monitoring and analytics. Weather data is collected automatically from the OpenWeatherMap API at regular intervals and processed through a simulated event- driven pipeline inspired by technologies such as Amazon Kinesis, AWS Lambda, and Amazon DynamoDB. Each weather record is treated as a streaming event and processed through ingestion, transformation, storage, and visualization layers. The backend system developed using Node.js manages data fetching, event processing, and communication with the frontend. The frontend dashboard built using React.js provides interactive visualization of weather parameters including temperature, humidity, pressure, and wind speed. WebSocket communication allows real-time updates without requiring manual page refresh. The proposed system demonstrates how real-time streaming architectures can support efficient analytics platforms and improve the responsiveness of weather monitoring applications.
Keywords: Real-Time Analytics, Weather Monitoring, Streaming Architecture, WebSockets, Cloud Computing
Keywords: Real-Time Analytics, Weather Monitoring, Streaming Architecture, WebSockets, Cloud Computing
Abstract
AWS BATCH FOR BATCH PROCESSING OF DOCKER CONTAINER
P. Kishan, Dr . S. Thavamani
DOI: 10.17148/IJARCCE.2026.15328
Abstract: Cloud computing has significantly transformed the way large-scale computational workloads are processed and managed. Traditional systems often struggle to efficiently handle highvolume batch jobs due to limitations in scalability, resource allocation, and automation. This paper presents a scalable solution for batch processing using AWS Batch integrated with containerized applications built using Docker.
The proposed system enables developers to run large numbers of batch computing jobs without manually provisioning or managing infrastructure. AWS Batch dynamically provisions the optimal quantity and type of compute resources based on the volume and requirements of submitted jobs. By packaging applications and their dependencies into Docker containers, the system ensures consistent execution environments, improved portability, and simplified deployment across cloud infrastructures.
In this project, Docker containers are stored and managed through Amazon Elastic Container Registry (ECR) and executed on scalable compute resources such as Amazon EC2 instances. AWS Batch automatically handles job scheduling, queue management, resource allocation, and monitoring, enabling efficient batch execution for workloads such as data processing, scientific simulations, and large-scale analytics.
The implemented architecture demonstrates how containerized workloads can be efficiently orchestrated using AWS Batch to improve performance, reduce operational complexity, and optimize resource utilization. The system also supports automated scaling and fault-tolerant execution, ensuring reliable job completion even under heavy workloads.
Overall, this research highlights the advantages of integrating AWS Batch with Docker-based containerization to build a robust, scalable, and cost-effective batch processing framework suitable for modern cloud-based applications.
The proposed system enables developers to run large numbers of batch computing jobs without manually provisioning or managing infrastructure. AWS Batch dynamically provisions the optimal quantity and type of compute resources based on the volume and requirements of submitted jobs. By packaging applications and their dependencies into Docker containers, the system ensures consistent execution environments, improved portability, and simplified deployment across cloud infrastructures.
In this project, Docker containers are stored and managed through Amazon Elastic Container Registry (ECR) and executed on scalable compute resources such as Amazon EC2 instances. AWS Batch automatically handles job scheduling, queue management, resource allocation, and monitoring, enabling efficient batch execution for workloads such as data processing, scientific simulations, and large-scale analytics.
The implemented architecture demonstrates how containerized workloads can be efficiently orchestrated using AWS Batch to improve performance, reduce operational complexity, and optimize resource utilization. The system also supports automated scaling and fault-tolerant execution, ensuring reliable job completion even under heavy workloads.
Overall, this research highlights the advantages of integrating AWS Batch with Docker-based containerization to build a robust, scalable, and cost-effective batch processing framework suitable for modern cloud-based applications.
Abstract
AWS LAMBDA FOR EVENT-DRIVEN MICROSERVICES ARCHITECTURE
Durga Nandhini BJ, Dr.S. Shylaja
DOI: 10.17148/IJARCCE.2026.15329
Abstract: Cloud computing has enabled organizations to build scalable and efficient systems without managing complex infrastructure. Event-driven microservices architecture is a modern approach that allows applications to react to events and process data dynamically. This project demonstrates the implementation of an event-driven microservices architecture using serverless technologies. The system uses Amazon S3 for file storage and event generation, while AWS Lambda processes the uploaded files automatically when an event occurs. Notifications are sent through Amazon SNS, and system monitoring is handled by Amazon CloudWatch. The architecture eliminates the need for server management and allows automatic scaling based on system demand. The proposed system improves efficiency, reduces operational overhead, and enables real-time processing of uploaded data. The results show that serverless event-driven architecture can provide a flexible, reliable, and scalable solution for modern cloud applications.
Keywords: Event-Driven Architecture, Serverless Computing, AWS Lambda, Microservices, Cloud Computing
Keywords: Event-Driven Architecture, Serverless Computing, AWS Lambda, Microservices, Cloud Computing
Abstract
DEPLOY A DOCKERIZED DJANGO APPLICATION ON AWS LIGHTSAIL
Kavya J, Dr.C.Daniel Nesakumar
DOI: 10.17148/IJARCCE.2026.15330
Abstract: In recent years, containerization and cloud computing have become important technologies for developing and deploying modern web applications. This project presents the deployment of a Dockerized Django web application called Finance Checker on AWS Lightsail. The main objective of this system is to help users manage and analyze their financial activities by tracking income and expenses through a web-based interface.
The application is developed using the Django framework, which provides a secure and scalable environment for building web applications. Docker is used to containerize the application along with all its dependencies, ensuring consistency across different environments and simplifying the deployment process. The Docker image is then deployed on AWS Lightsail, a cloud computing platform that offers easy and cost-effective hosting for applications.
Keywords: Docker, Django Framework, AWS Lightsail, Cloud Computing, Containerization, Web Application Deployment, Finance Checker, Docker Containers, Cloud Hosting, Financial Data Management.
The application is developed using the Django framework, which provides a secure and scalable environment for building web applications. Docker is used to containerize the application along with all its dependencies, ensuring consistency across different environments and simplifying the deployment process. The Docker image is then deployed on AWS Lightsail, a cloud computing platform that offers easy and cost-effective hosting for applications.
Keywords: Docker, Django Framework, AWS Lightsail, Cloud Computing, Containerization, Web Application Deployment, Finance Checker, Docker Containers, Cloud Hosting, Financial Data Management.
Abstract
MULTI-STAGE DOCKER BUILDS AND DEPLOYMENT ON AWS ECS
Nandhini S, Dr. B. Narasimhan
DOI: 10.17148/IJARCCE.2026.15331
Abstract: Modern software applications require scalable, portable, and efficient deployment mechanisms. Traditional deployment methods often involve manual configuration of servers and dependencies, which increases complexity and deployment time. Containerization technologies such as Docker address these issues by packaging applications along with their dependencies into lightweight containers that can run consistently across different environments. However, single-stage container builds may create large images containing unnecessary dependencies.
This project presents a cloud-native deployment architecture that uses multi-stage Docker builds to create optimized container images and deploy them using Amazon Elastic Container Service (ECS). The application is containerized using Docker, stored in a container registry, and executed in a serverless container environment using AWS Fargate. The system also integrates monitoring and logging capabilities through centralized logging services to observe runtime performance and system behavior.
The proposed architecture demonstrates how optimized container images can be efficiently deployed and managed in a cloud environment while ensuring scalability, reliability, and security.
Keywords: Docker, Multi-Stage Builds, Containerization, AWS ECS, Cloud Deployment, DevOps
This project presents a cloud-native deployment architecture that uses multi-stage Docker builds to create optimized container images and deploy them using Amazon Elastic Container Service (ECS). The application is containerized using Docker, stored in a container registry, and executed in a serverless container environment using AWS Fargate. The system also integrates monitoring and logging capabilities through centralized logging services to observe runtime performance and system behavior.
The proposed architecture demonstrates how optimized container images can be efficiently deployed and managed in a cloud environment while ensuring scalability, reliability, and security.
Keywords: Docker, Multi-Stage Builds, Containerization, AWS ECS, Cloud Deployment, DevOps
Abstract
Eye Movement Driven Cursor Control Using Computer Vision Technique
Karthik. V, Dr. M.Hemalatha, M.Sc., M.Phil., Ph.D
DOI: 10.17148/IJARCCE.2026.15332
Abstract: In the field of human-computer interaction, eye tracking technology has drawn a lot of attention, especially for helping people with physical disabilities who can't use conventional input devices like a mouse or keyboard. The creation of an inexpensive eye-controlled mouse system that allows users to manipulate the movement of a computer cursor by blinking and eye gaze is presented in this paper. Using a standard webcam, the proposed system tracks eye movements in real time and detects facial landmarks using computer vision and deep learning techniques. Eye positions are determined using image processing techniques, and cursor control is achieved by mapping gaze direction to screen coordinates. Click operations are carried out by blink detection, enabling users to engage with applications without making physical contact. Python is used to implement the system, and libraries like OpenCV, MediaPipe, and PyAutoGUI are used for cursor automation, video processing, and facial landmark detection. The suggested method can provide precise cursor movement and dependable click detection under typical lighting conditions, according to experimental results. For users with mobility impairments, the developed system provides an accessible and reasonably priced assistive technology solution that improves computer accessibility and hands-free interaction.
Abstract
Implement Serverless Machine Learning Inference with AWS Lambda
Abishek P, Dr. T. Pradeep
DOI: 10.17148/IJARCCE.2026.15333
Abstract: Serverless computing has emerged as a transformative paradigm for deploying machine learning models at scale without infrastructure management overhead. Traditional deployment approaches require dedicated servers with fixed capacity and 24/7 operational costs, creating barriers for machine learning adoption in cost-sensitive applications. This paper presents the implementation of a serverless machine learning inference system using AWS Lambda, demonstrating its application in intelligent loan approval decisions. The system implements a complete end-to-end architecture combining modern web technologies (React.js, FastAPI, MongoDB) with AWS serverless services (Lambda, API Gateway, DynamoDB) to deliver real-time ML inference with automatic scaling and pay-per-use economics. The core contribution is a transparent AI risk evaluation engine that analyzes five financial parameters— Debt Service Ratio, Loan-to-Income Ratio, credit score, employment stability, and dependent count—to compute composite risk scores (0- 100) and generate instant loan decisions (APPROVED, CONDITIONAL, REJECTED) with complete explainability. The serverless architecture eliminates infrastructure provisioning, enables automatic scaling from zero to thousands of concurrent requests, and reduces operational costs by 70% compared to traditional always-on server deployments. Performance evaluation demonstrates average inference response times under 3 seconds, System Usability Scale score of 83.5 (Excellent), and zero security vulnerabilities across comprehensive penetration testing. The system processes loan applications 80x faster than traditional manual processes while maintaining complete transparency and eliminating human bias through deterministic algorithms. This implementation validates serverless computing as a viable, cost- effective architecture for deploying production-grade machine learning inference systems in financial services and demonstrates practical patterns for Lambda function development, API Gateway integration, and DynamoDB optimization.
Keywords: Serverless Computing, AWS Lambda, Machine Learning Inference, Real-time Prediction, API Gateway, DynamoDB, FastAPI, React, Cloud Architecture, Scalable ML Deployment.
Keywords: Serverless Computing, AWS Lambda, Machine Learning Inference, Real-time Prediction, API Gateway, DynamoDB, FastAPI, React, Cloud Architecture, Scalable ML Deployment.
Abstract
Optimization of Niche Retail Operations: Design and Implementation of the "Deera Store Management System" web application
Asif reehan J, Dr. G. Maria Priscilla
DOI: 10.17148/IJARCCE.2026.15334
Abstract: In the specialized sector of foreign goods retail, efficient inventory and sales management are paramount due to the diversity of products and supply chain complexities. This paper presents the design and implementation of the Deera Store Management System, a desktop-based web application developed specifically for "Deera Shopping and Foreign Goods." The system addresses the limitations of manual ledger keeping by integrating inventory monitoring, sales tracking, and supplier management into a unified platform. Built using a hybrid technology stack comprising node, HTML, CSS, and JavaScript, the application streamlines daily operations, provides low-stock alerts, and offers real-time analytics. The deployment of this system demonstrates a significant reduction in data inconsistency and an improvement in operational efficiency and trend identification.
Keywords: Retail Automation, Inventory Management System (IMS), Node.js, Foreign Goods, Sales Analytics, Desktop-Web Hybrid Application.
Keywords: Retail Automation, Inventory Management System (IMS), Node.js, Foreign Goods, Sales Analytics, Desktop-Web Hybrid Application.
Abstract
Emotion-Aware AI for Enhancing Elderly Mental Health
Abhishek R Shetty, Achith J V, Adithya Jain, Aman N, Mr. Shivaraj B G
DOI: 10.17148/IJARCCE.2026.15335
Abstract: Emotion recognition plays an important role in understanding the mental and emotional well-being of individuals, especially in elderly care. With the rapid growth of the aging population, there is an increasing need for automated systems that can continuously monitor emotional states and provide timely support. This project presents an emotion-aware artificial intelligence system that detects human emotions from facial expressions using deep learning techniques. The system analyzes facial images uploaded by users and classifies emotions such as happy, sad, angry, fearful, surprised, and neutral. A customized Convolutional Neural Network (CNN) model is used for emotion detection due to its ability to automatically extract discriminative facial features and handle variations in facial expressions. The application is developed using the Django framework for backend operations and a relational database for storing user information, emotion records, appointments, and reminders. The system also integrates healthcare support features such as doctor appointment scheduling, medicine reminders, and WhatsApp-based notifications to caregivers and doctors. Experimental results indicate that the proposed model achieves reliable emotion classification performance and supports timely emotional monitoring, thereby enhancing mental health care and quality of life for elderly individuals.
Keywords: Emotion Recognition, Facial Expression Analysis, Deep Learning, Convolutional Neural Network, Elderly Mental Health, Computer Vision, Healthcare Monitoring.
Keywords: Emotion Recognition, Facial Expression Analysis, Deep Learning, Convolutional Neural Network, Elderly Mental Health, Computer Vision, Healthcare Monitoring.
Abstract
Attention-Based Occlusion Handling for Dynamic Human-Robot Interaction in Unstructured Environments
Muthuraj S, Dr. S. Mohana
DOI: 10.17148/IJARCCE.2026.15336
Abstract: Robust human detection and pose estimation under occlusion remain central unsolved challenges for robots operating in unstructured, cluttered, or dynamically evolving environments. Conventional perception pipelines degrade substantially when human body parts are partially or fully occluded by furniture, equipment, or other people, leading to unstable robot behaviour and potentially unsafe interactions. This paper proposes the Attention-Gated Occlusion- Robust Perception (AGORP) framework, a novel architecture that integrates spatial and channel attention with transformer-based contextual reasoning to enable real-time occlusion detection, body-part recovery, and intention prediction during human-robot interaction. The framework introduces a dual-stream encoder that processes RGB and depth modalities through cross-modal attention gates, coupled with an Occlusion-Aware Temporal Transformer (OATT) that exploits motion history to hallucinate occluded skeletal configurations. Experiments conducted on the JRDB, CrowdBot, and a newly collected indoor manipulation dataset demonstrate that AGORP achieves a mean Average Precision (mAP) of 74.3% on occluded pose estimation, a 12.7 percentage-point improvement over the strongest baseline, while sustaining 31 frames per second on an NVIDIA Jetson AGX Orin embedded platform. Qualitative and quantitative analyses confirm the framework’s generalisation across occlusion severities, scene densities, and lighting conditions.
Keywords: attention mechanism, occlusion handling, human-robot interaction, transformer networks, pose estimation, unstructured environments.
Keywords: attention mechanism, occlusion handling, human-robot interaction, transformer networks, pose estimation, unstructured environments.
Abstract
ML - Based Traffic Violation Severity Scorer
S. Ratheesh Kumar, Dr. N. Mahendiran
DOI: 10.17148/IJARCCE.2026.15337
Abstract: The title of this project is "Ml-Based Traffic Violation Security Score.” This project is developed using HTML, CSS, Bootstrap, and JS for the front end, along with a Python back end, where the framework is Flask.This project is based on the evaluation and prediction of the risk of accidents happening on the road. It is based on the analysis of the history of accidents, along with detailed information regarding the drivers, vehicles, and violations, to evaluate the potential hazards that may occur. The features that are considered for the prediction of the potential causes of accidents include Belts, Personal Injury, Property Damage, Commercial License, Commercial Vehicle, Contributed To Accident, Vehicle Type, Year, Make, Model, Color, Charge, Driver City, Driver State, Arrest Type, and Violation Type. It uses an algorithm that can handle the features efficiently for the prediction results. On the basis of the obtained results, it classifies the level of risk as Low, Medium, or High. This will help you decide how to use your resources to issue warnings and prevent accidents before they happen. It uses real data to help you make smart decisions and keep the roads safe by preventing accidents. It’s a combination of past data, predictions, smart scoring, and an efficient system that can be used to manage traffic.
Keywords: Machine Learning, Traffic Violation Prediction, Road Safety Analysis, Random Forest Algorithm, Traffic Risk Assessment, Traffic Violation Security Score, Accident Prediction, Intelligent Traffic Management.
Keywords: Machine Learning, Traffic Violation Prediction, Road Safety Analysis, Random Forest Algorithm, Traffic Risk Assessment, Traffic Violation Security Score, Accident Prediction, Intelligent Traffic Management.
Abstract
REAL-TIME LOG ANALYSIS WITH AWS OPENSEARCH AND DOCKER
Abinaya G, Mr. T. Pradeep
DOI: 10.17148/IJARCCE.2026.15338
Abstract: Cloud computing environments generate a massive amount of operational and security logs due to continuous API calls, user authentication activities, and service interactions. Monitoring these logs manually is inefficient and cannot provide real-time insights for security and operational management. This research presents a centralized real-time log monitoring system using AWS CloudTrail, Docker, and Amazon OpenSearch. The proposed system automatically collects CloudTrail logs stored in Amazon S3, processes them through a Dockerbased log ingestion engine, and indexes them in Amazon OpenSearch for fast search and analysis.
OpenSearch Dashboards provide visualization tools that help administrators detect suspicious activities, analyze system usage patterns, and improve cloud security monitoring. The system ensures automation, scalability, centralized visibility, and efficient log analytics in AWS environments. By integrating AWS services with containerization technologies, the system simplifies log monitoring and enables faster incident detection. The solution provides an effective framework for real-time monitoring of cloud infrastructure activities.
Keywords: CloudTrail, OpenSearch, Docker, Log Analysis, Cloud Security, AWS
OpenSearch Dashboards provide visualization tools that help administrators detect suspicious activities, analyze system usage patterns, and improve cloud security monitoring. The system ensures automation, scalability, centralized visibility, and efficient log analytics in AWS environments. By integrating AWS services with containerization technologies, the system simplifies log monitoring and enables faster incident detection. The solution provides an effective framework for real-time monitoring of cloud infrastructure activities.
Keywords: CloudTrail, OpenSearch, Docker, Log Analysis, Cloud Security, AWS
Abstract
Sub-Pixel Semantic Segmentation for Precision Agriculture: Identifying Micro-Defects in Crop Foliage
Akshay K, Dr. S. Devibala Subramanian
DOI: 10.17148/IJARCCE.2026.15339
Abstract: Early detection of micro-defects in crop foliage—including sub-millimetre chlorotic lesions, fungal hyphae boundaries, and early-stage necrotic patches—remains one of the most challenging problems in computational precision agriculture. Conventional semantic segmentation models operate at the native pixel resolution of remotely-sensed or macro-lens imagery and consistently fail to recover boundary-level structural detail at scales below one pixel. This paper presents Sub-Pixel Segmentation Network (SPSNet), a novel deep-learning architecture that incorporates learned sub- pixel convolution layers, spectral-channel attention, and a multi-scale hierarchical decoder to segment foliage defects at resolutions exceeding the sensor's native sampling grid. Our model is trained and evaluated on FoliageDefect-22K, a purpose-built annotated dataset comprising 22,400 high-resolution images across six crop species and nine defect categories collected under controlled and field conditions. SPSNet achieves a mean Intersection-over-Union (mIoU) of 91.7%, surpassing the nearest competing method by 8.5 percentage points, while maintaining near-real-time inference at 18.3 frames per second on a single NVIDIA A100 GPU. Ablation studies confirm the individual contribution of each architectural component. These results establish sub-pixel segmentation as a viable and practical tool for deployment in precision agricultural monitoring systems.
Keywords: precision agriculture, semantic segmentation, sub-pixel convolution, foliage micro-defects, deep learning, crop health monitoring, spectral attention
Keywords: precision agriculture, semantic segmentation, sub-pixel convolution, foliage micro-defects, deep learning, crop health monitoring, spectral attention
Abstract
Cybersecurity threat detection to Prevent System
Tikshanaa.P.R, Abinaya.Y, Dr.R.Nagarajan
DOI: 10.17148/IJARCCE.2026.15340
Abstract: Cybersecurity Threat Detection is a system designed to help detect and prevent cyber-attacks in real time. In today’s digital world, hackers attempt to steal sensitive data, disrupt services, or damage systems using methods such as malware, phishing, ransomware, and unauthorized access. With the rapid growth of online transactions, cloud computing, and interconnected devices, the risk of cyber threats has increased significantly. This project aims to protect users and organizations by identifying these threats early, generating alerts, and enabling quick defensive action. The system continuously monitors network activity, user behaviour, and system logs to identify anomalies. When a possible threat is detected, it sends an immediate warning and records detailed information for further analysis. Detection techniques include signature-based scanning for known attack patterns, anomaly detection to spot unusual traffic or login attempts, and behavioural analysis to identify suspicious actions from users or devices. By combining these approaches, the system achieves a balance between accuracy and efficiency, reducing false positives while ensuring that genuine threats are not overlooked. The project is developed using HTML, CSS, and JavaScript for the front-end interface, providing a user- friendly dashboard where alerts and logs can be viewed clearly. The Python-based back-end handles data processing, threat detection algorithms, and communication between system components. The dashboard allows administrators to track incidents, analyse patterns, and take corrective measures such as blocking IP addresses, disabling compromised accounts, or applying patches. This system contributes to reducing the risk of cyber-attacks by offering proactive monitoring and timely alerts. It empowers users to safeguard their data, maintain system integrity, and comply with security standards. Beyond individual protection, the project demonstrates how organizations can adopt layered security measures to strengthen resilience against evolving cyber threats. Ultimately, Cybersecurity Threat Detection enhances trust in digital platforms and supports the safe growth of technology-driven environments.
Keywords: Cybersecurity, Threat Detection, Intrusion Detection System (IDS), Anomaly Detection, Malware Prevention, Network Security, Real-Time Monitoring, Incident Response.
Keywords: Cybersecurity, Threat Detection, Intrusion Detection System (IDS), Anomaly Detection, Malware Prevention, Network Security, Real-Time Monitoring, Incident Response.
Abstract
ADAPTIVE CLOUD LEARNING TUTOR USING LLM
Prince Rosario, Shriharri, Ms A Deepika
DOI: 10.17148/IJARCCE.2026.15341
Abstract: So these are the contents of my research paper. These are some PPT slides I want you to convert into a paper presentation which consists of six pages. I want you to give me abstract, keywords, introduction, proposed system architecture, methodology with some modules, and experiments and results, results and discussions, conclusions, and some references. Please don't add on your own. If you want any other information, you can ask me. Make it plagiarism- free, more humanized way. That's it. Make it in a structured manner. Learning Management Systems (LMS) have become an essential component of modern education, enabling institutions to deliver courses, manage assessments, and track learner progress through digital platforms. However, most traditional LMS platforms operate using static recommendation techniques and predefined rules, offering minimal personalization to learners. As a result, students often struggle to identify suitable courses aligned with their interests, skill levels, and academic goals. Additionally, mentor allocation is frequently manual or generic, reducing the effectiveness of guidance and academic support. This research proposes an Intelligent Learning Management System (iLMS) powered by Large Language Models (LLMs) and collaborative recommendation mechanisms. The system processes structured and unstructured data such as student interactions, course materials, feedback, and mentor profiles to generate semantic embeddings. These embeddings capture learner preferences, knowledge levels, engagement patterns, and mentor expertise. Based on this understanding, the system delivers personalized course recommendations, adaptive learning paths, and intelligent mentor-student matching. The proposed system also integrates real-time performance monitoring, predictive analytics, and dynamic assessment mechanisms to identify knowledge gaps and adjust content difficulty accordingly. Interactive dashboards provide continuous feedback to students, mentors, and administrators. By combining LLM-based semantic analysis with adaptive recommendation strategies, the iLMS enhances learner engagement, improves academic outcomes, and provides a scalable solution for personalized digital education.
Keywords: Learning Management System, Large Language Model (LLM), Personalized Learning, Course Recommendation, Adaptive Learning Path, Mentor-Student Matching, Semantic Embedding, NLP, Collaborative Recommendation, Educational Data Mining.
Keywords: Learning Management System, Large Language Model (LLM), Personalized Learning, Course Recommendation, Adaptive Learning Path, Mentor-Student Matching, Semantic Embedding, NLP, Collaborative Recommendation, Educational Data Mining.
Abstract
SERVERLESS IMAGE PROCESSING WITH
AWS LAMBDA AND S, Rithika. P, Dr. S. Shylaja
DOI: 10.17148/IJARCCE.2026.15343
Abstract: Cloud computing provides scalable, flexible, and cost-effective solutions for modern applications. This paper presents a serverless image processing system developed using Amazon Web Services (AWS). In the proposed system, users upload images to Amazon S3, which acts as cloud storage. When an image is uploaded, an event notification automatically triggers an AWS Lambda function that performs the required image processing operation. The processed images are then stored in a separate output bucket for easy retrieval. Amazon CloudWatch is used to monitor system activity and maintain execution logs. The serverless architecture eliminates the need for managing servers or infrastructure, allowing the system to scale automatically based on demand. This approach improves efficiency, reduces operational cost, and simplifies the deployment of cloud-based image processing applications.
Keywords: Cloud Computing, Serverless Architecture, AWS Lambda, Amazon S3, Image Processing, Cloud Storage.
Keywords: Cloud Computing, Serverless Architecture, AWS Lambda, Amazon S3, Image Processing, Cloud Storage.
Abstract
Secure docker container with AWS secrets manager
M.J.Premkarthik, Dr. S. Shylaja
DOI: 10.17148/IJARCCE.2026.15344
Abstract: Modern software systems require flexible and scalable deployment environments to support growing workloads and complex applications. Containerization has emerged as an efficient solution that allows applications to be packaged with all required dependencies into lightweight containers. Docker is one of the most widely used containerization technologies that enables developers to build portable and consistent application environments. However, managing sensitive data such as API keys, database passwords, and configuration tokens within these containers requires a highly secure and reliable approach.
This research presents a cloud-based security architecture titled “Secure Docker Container with AWS Secrets Manager.” The system integrates Docker containerization with Amazon Web Services (AWS) security components, primarily focusing on AWS Secrets Manager, along with Amazon Elastic Compute Cloud (EC2) and AWS Identity and Access Management (IAM).
The proposed system demonstrates a highly secure container deployment model that improves application safety, credential management, and compliance. The results show that combining Docker with AWS Secrets Manager can significantly reduce security vulnerabilities and support modern DevSecOps practices for building resilient, secure applications.
Keywords: Docker, AWS Secrets Manager, Container Security, Cloud Deployment, Credential Management, DevSecOps.
This research presents a cloud-based security architecture titled “Secure Docker Container with AWS Secrets Manager.” The system integrates Docker containerization with Amazon Web Services (AWS) security components, primarily focusing on AWS Secrets Manager, along with Amazon Elastic Compute Cloud (EC2) and AWS Identity and Access Management (IAM).
The proposed system demonstrates a highly secure container deployment model that improves application safety, credential management, and compliance. The results show that combining Docker with AWS Secrets Manager can significantly reduce security vulnerabilities and support modern DevSecOps practices for building resilient, secure applications.
Keywords: Docker, AWS Secrets Manager, Container Security, Cloud Deployment, Credential Management, DevSecOps.
Abstract
DOCKER SWARM CLUSTER ON AWS
A. Taniya Maria, Dr. C. Daniel Nesakumar
DOI: 10.17148/IJARCCE.2026.15345
Abstract: Docker Swarm is a native container orchestration platform that enables the deployment, management, and scaling of containerized applications across multiple machines. Implementing a Docker Swarm cluster on Amazon Web Services (AWS) provides a scalable, highly available, and fault-tolerant environment for modern cloud applications. In this architecture, multiple Amazon EC2 instances are configured as nodes within the swarm cluster, where manager nodes handle cluster management tasks such as scheduling, orchestration, and maintaining the desired state of services, while worker nodes run the application containers. AWS infrastructure components such as Elastic Compute Cloud (EC2), Elastic Load Balancer (ELB), Elastic Block Store (EBS), and Elastic File System (EFS) are integrated to support compute resources, traffic distribution, and persistent storage. The cluster is typically distributed across multiple availability zones to ensure high availability and resilience against failures. Docker Swarm simplifies container networking through overlay networks and provides built-in load balancing and service discovery. By deploying Docker Swarm on AWS, organizations can efficiently manage microservices-based applications, automate container deployment, and dynamically scale services according to workload demands while leveraging the reliability, security, and elasticity of the AWS cloud platform.
Abstract
“AI-Based Therapist Chatbot for Emotional Support”
Santhiya.M, Ms. A. Deepika
DOI: 10.17148/IJARCCE.2026.15346
Abstract: Mental health problems like stress, anxiety, and sadness are common today. Many people do not talk about their problems because they feel shy or do not have access to a therapist. This paper presents an AI-based therapist chatbot that provides emotional support to users. The chatbot uses Natural Language Processing (NLP) to understand what the user is saying and give helpful replies. When a user types their feelings, the chatbot analyses the text and responds with supportive messages. The system helps users share their emotions and feel better. This chatbot does not replace professional doctors, but it can give basic emotional support anytime.
Keywords: Artificial Intelligence, Chatbot, Emotional Support, NLP, Mental Health
Keywords: Artificial Intelligence, Chatbot, Emotional Support, NLP, Mental Health
Abstract
Implementing Infrastructure Monitoring Using Docker, Prometheus and Grafana
P. Merlyn Jessica, Dr. B. Narasimhan
DOI: 10.17148/IJARCCE.2026.15347
Abstract: Modern IT infrastructures require continuous monitoring to ensure system reliability, performance, and scalability. Traditional monitoring solutions often involve complex configurations and limited scalability. This paper presents an efficient infrastructure monitoring system implemented using containerization and open-source monitoring tools, namely Docker, Prometheus, and Grafana. Docker is used to create lightweight and portable containers for deploying monitoring services. Prometheus collects real-time system metrics such as CPU usage, memory utilization, disk performance, and network activity. Grafana provides an interactive dashboard for visualization and analysis of collected metrics. The proposed system enables administrators to track infrastructure performance, detect anomalies, and maintain system stability through automated monitoring and graphical insights. The results demonstrate that the integration of containerization with modern monitoring tools provides a flexible, scalable, and efficient monitoring framework suitable for cloud environments and enterprise systems.
Keywords: Infrastructure Monitoring, Docker, Prometheus, Grafana, Containerization, System Performance Monitoring.
Keywords: Infrastructure Monitoring, Docker, Prometheus, Grafana, Containerization, System Performance Monitoring.
Abstract
SECURE AWS INFRASTRUCTURE WITH DOCKER AND IAM ROLES
Sananta.M, Dr. B. Narasimhan
DOI: 10.17148/IJARCCE.2026.15348
Abstract: The project, Secure AWS Infrastructure Using Docker and IAM Roles, focuses on implementing a secure and efficient infrastructure for deploying applications using services provided by Amazon Web Services. The application is deployed on a virtual server using Amazon EC2 and runs inside a containerized environment using Docker. Containerization ensures that the application and its dependencies are packaged together, enabling consistent execution and simplified deployment.
To enhance security, the project utilizes role-based access control through AWS Identity and Access Management. IAM roles allow the application running on the EC2 instance to securely access other AWS services without storing permanent credentials in the source code. This approach improves security by providing temporary and controlled permissions.
The system stores and retrieves application data using Amazon S3, which provides reliable and scalable storage. The application interacts with S3 using the AWS SDK, enabling secure communication between the application and storage service. Overall, the project demonstrates how containerization and role-based access control can be integrated to build a secure and manageable AWS infrastructure for modern application deployment.
Keywords: AWS Infrastructure, Docker, IAM Roles, Amazon EC2, Amazon S3, Containerization, AWS SDK.
To enhance security, the project utilizes role-based access control through AWS Identity and Access Management. IAM roles allow the application running on the EC2 instance to securely access other AWS services without storing permanent credentials in the source code. This approach improves security by providing temporary and controlled permissions.
The system stores and retrieves application data using Amazon S3, which provides reliable and scalable storage. The application interacts with S3 using the AWS SDK, enabling secure communication between the application and storage service. Overall, the project demonstrates how containerization and role-based access control can be integrated to build a secure and manageable AWS infrastructure for modern application deployment.
Keywords: AWS Infrastructure, Docker, IAM Roles, Amazon EC2, Amazon S3, Containerization, AWS SDK.
Abstract
Automate Docker Container Management with AWS Elastic Beanstalk
Dharaneeswaran R B, Ms. A. Praveena
DOI: 10.17148/IJARCCE.2026.15349
Abstract: Modern software systems require flexible and scalable deployment environments to support growing workloads and complex applications. Containerization has emerged as an efficient solution that allows applications to be packaged with all required dependencies into lightweight containers. Docker is one of the most widely used containerization technologies that enables developers to build portable and consistent application environments. However, managing containers in production requires an automated environment that handles provisioning and load balancing.
This research presents a cloud-based deployment architecture titled "Automated Docker Container Management with AWS Elastic Beanstalk." The system integrates Docker containerization with several Amazon Web Services (AWS) components including AWS Elastic Beanstalk, Amazon Elastic Container Registry (ECR), and AWS Identity and Access Management (IAM). Docker is used to build container images, while Amazon ECR stores these images securely. AWS Elastic Beanstalk acts as the orchestration layer, automatically handling the deployment details of capacity provisioning, load balancing, auto-scaling, and application health monitoring.
The proposed system demonstrates a streamlined container deployment model that improves application portability and operational efficiency. The results show that leveraging AWS Elastic Beanstalk for Docker management significantly simplifies the transition from local development to cloud-scale production while supporting modern DevOps practices.
Keywords: Docker, AWS Elastic Beanstalk, Cloud Deployment, Containerization, Automation, PaaS, DevOps.
This research presents a cloud-based deployment architecture titled "Automated Docker Container Management with AWS Elastic Beanstalk." The system integrates Docker containerization with several Amazon Web Services (AWS) components including AWS Elastic Beanstalk, Amazon Elastic Container Registry (ECR), and AWS Identity and Access Management (IAM). Docker is used to build container images, while Amazon ECR stores these images securely. AWS Elastic Beanstalk acts as the orchestration layer, automatically handling the deployment details of capacity provisioning, load balancing, auto-scaling, and application health monitoring.
The proposed system demonstrates a streamlined container deployment model that improves application portability and operational efficiency. The results show that leveraging AWS Elastic Beanstalk for Docker management significantly simplifies the transition from local development to cloud-scale production while supporting modern DevOps practices.
Keywords: Docker, AWS Elastic Beanstalk, Cloud Deployment, Containerization, Automation, PaaS, DevOps.
Abstract
Serverless ELT Pipeline for Scalable Data Processing Using AWS Glue and AWS Lambda
Isacc Moses S, Mr. Daniel Nesa Kumar C
DOI: 10.17148/IJARCCE.2026.15350
Abstract: The Serverless ELT (Extract, Load, Transform) pipeline leveraging AWS Glue and AWS Lambda offers a modern, fully managed approach for handling large-scale data processing workloads. By adopting a serverless architecture, this solution eliminates the need for manual server provisioning, configuration, and maintenance, allowing organizations to focus solely on data operations and insights. AWS Lambda serves as the orchestration layer, automating the triggering of data ingestion workflows from multiple sources including relational databases, APIs, and object storage systems. This event-driven execution ensures that data pipelines operate efficiently, respond to changes in real-time, and scale seamlessly without manual intervention. Once ingested, data is loaded into Amazon S3, forming a centralized, durable, and secure data lake. AWS Glue then handles automated data cataloging, schema discovery, and transformation using its distributed Apache Spark-based processing engine. The serverless nature of this pipeline provides significant operational and cost advantages through pay-as-you-go pricing, automated logging, monitoring, and error handling features that reduce operational overhead and enhance system reliability.
Keywords: AWS Lambda, AWS Glue, Serverless Architecture, ELT Pipeline, Amazon S3, Cloud Computing, Data Transformation, Apache Spark, Amazon CloudWatch, Python.
Keywords: AWS Lambda, AWS Glue, Serverless Architecture, ELT Pipeline, Amazon S3, Cloud Computing, Data Transformation, Apache Spark, Amazon CloudWatch, Python.
Abstract
Serverless File Processing Pipeline with AWS Step Functions
Sampathkumar R S, Dr. T. Pradeep
DOI: 10.17148/IJARCCE.2026.15351
Abstract: The Serverless File Processing Pipeline with AWS Step Functions is a cloud-based automated system designed to process uploaded files efficiently without managing physical servers. This project demonstrates how modern cloud technologies can automate document processing using a serverless architecture. When a file such as a resume or document is uploaded to Amazon S3, an event is generated that automatically triggers a workflow managed by AWS Step Functions. The workflow coordinates multiple AWS Lambda functions to perform sequential tasks such as file validation, text extraction, and data storage. The extracted content is stored in an output S3 bucket, while metadata and processing details are saved in Amazon DynamoDB for record management. Upon successful completion, a notification email is sent to the user through Amazon SNS. The architecture is event-driven, cost-effective, secure, and highly reliable. The project demonstrates real-world implementation of cloud computing concepts such as workflow orchestration, event handling, automated processing, and serverless design, providing an efficient and modern solution for automated file processing.
Keywords: Serverless Architecture, AWS Step Functions, AWS Lambda, Amazon S3, DynamoDB, Amazon SNS, Cloud Computing, Event-Driven Architecture, File Processing, Automation.
Keywords: Serverless Architecture, AWS Step Functions, AWS Lambda, Amazon S3, DynamoDB, Amazon SNS, Cloud Computing, Event-Driven Architecture, File Processing, Automation.
Abstract
Deploying a WordPress Site Using Docker and AWS RDS
S. Pranesh, Dr. S. Thavamani
DOI: 10.17148/IJARCCE.2026.15352
Abstract: In modern cloud computing environments, deploying scalable and reliable web applications is essential for ensuring performance, availability, and efficient resource management. Traditional web hosting methods often involve manual configuration and dependency management, which can lead to inconsistencies across deployment environments. Containerization technologies and cloud services provide a solution to these challenges by enabling automated deployment, portability, and scalability.
This paper presents the deployment of a WordPress website using Docker containerization on an Amazon EC2 instance integrated with Amazon Relational Database Service (RDS) for database management. WordPress is a widely used Content Management System (CMS) that allows users to create and manage dynamic websites. By deploying WordPress inside a Docker container, the application environment becomes isolated, portable, and easy to maintain. Amazon EC2 provides scalable cloud infrastructure to host the application, while Amazon RDS offers a managed MySQL database service for secure and reliable data storage.
The proposed architecture follows a two-tier model consisting of an application layer and a database layer. The application layer hosts the Docker container running WordPress on an EC2 instance, while the database layer uses Amazon RDS to store website content and user data. Security groups and network configurations ensure secure communication between the two components. Elastic IP is used to provide consistent external access to the hosted website.
This paper presents the deployment of a WordPress website using Docker containerization on an Amazon EC2 instance integrated with Amazon Relational Database Service (RDS) for database management. WordPress is a widely used Content Management System (CMS) that allows users to create and manage dynamic websites. By deploying WordPress inside a Docker container, the application environment becomes isolated, portable, and easy to maintain. Amazon EC2 provides scalable cloud infrastructure to host the application, while Amazon RDS offers a managed MySQL database service for secure and reliable data storage.
The proposed architecture follows a two-tier model consisting of an application layer and a database layer. The application layer hosts the Docker container running WordPress on an EC2 instance, while the database layer uses Amazon RDS to store website content and user data. Security groups and network configurations ensure secure communication between the two components. Elastic IP is used to provide consistent external access to the hosted website.
Abstract
AWS CloudTrail for Monitoring Docker Container Activity
PRATEESH S, Mr. S. S. Saravana Kumar
DOI: 10.17148/IJARCCE.2026.15353
Abstract: With the rapid adoption of containerization technologies such as Docker, monitoring container activity and maintaining security has become an important concern in modern cloud environments. This paper proposes a monitoring framework using Amazon Web Services (AWS) CloudTrail to track and analyze Docker container activity. AWS CloudTrail records API calls and user activities across AWS infrastructure, enabling administrators to monitor system operations, detect anomalies, and ensure compliance. The proposed system integrates Docker container deployment on AWS infrastructure with CloudTrail logging mechanisms to capture container-related events and activities. The collected logs can be stored, analyzed, and visualized to provide insights into container behavior, security incidents, and operational performance. This approach enhances transparency and accountability in containerized environments and supports effective auditing and security management.
Abstract
Serverless Data Processing Using AWS Lambda and Amazon Kinesis
H Darsan, Mr. B. Ramesh Kumar
DOI: 10.17148/IJARCCE.2026.15354
Abstract: Cloud computing has significantly transformed the development and deployment of modern applications. One of the most important advancements in cloud technology is serverless computing, which enables developers to build scalable applications without managing server infrastructure. This paper presents a serverless data processing architecture using Amazon Web Services (AWS), specifically utilizing AWS Lambda and Amazon Kinesis Data Streams to process real-time e-commerce order data.
The proposed system integrates multiple AWS services including Amazon API Gateway, AWS Lambda, Amazon Kinesis Data Streams, and Amazon S3. The frontend application collects order information from users and sends it to an API endpoint. API Gateway triggers a Lambda function that sends the order data to a Kinesis stream. Another Lambda function consumes the streaming records, processes the order information, and stores the processed data in Amazon S3. The system demonstrates how serverless architecture can efficiently process streaming data while eliminating the need for server management. Experimental results show that the system can handle real-time data processing effectively while providing scalability, reliability, and cost efficiency. This study highlights the potential of serverless computing for building scalable cloud-based data processing systems.
Keywords: Serverless Computing, AWS Lambda, Amazon Kinesis, Cloud Computing, Real-Time Data Processing, Event-Driven Architecture.
The proposed system integrates multiple AWS services including Amazon API Gateway, AWS Lambda, Amazon Kinesis Data Streams, and Amazon S3. The frontend application collects order information from users and sends it to an API endpoint. API Gateway triggers a Lambda function that sends the order data to a Kinesis stream. Another Lambda function consumes the streaming records, processes the order information, and stores the processed data in Amazon S3. The system demonstrates how serverless architecture can efficiently process streaming data while eliminating the need for server management. Experimental results show that the system can handle real-time data processing effectively while providing scalability, reliability, and cost efficiency. This study highlights the potential of serverless computing for building scalable cloud-based data processing systems.
Keywords: Serverless Computing, AWS Lambda, Amazon Kinesis, Cloud Computing, Real-Time Data Processing, Event-Driven Architecture.
Abstract
Deploy a Dockerized E-commerce Platform on AWS ECH
Dinesh Kumar, Dr. C. Daneil Nesakumar
DOI: 10.17148/IJARCCE.2026.15355
Abstract: The increasing demand for scalable and reliable web applications has encouraged the adoption of cloud computing and containerization technologies. Traditional deployment methods often involve complex server management and dependency issues that affect system performance and scalability. This research presents the design and deployment of a Dockerized e-commerce platform using Amazon Web Services (AWS) Elastic Container Service (ECS). The proposed system demonstrates how containerization can simplify application deployment and improve consistency across different environments. The e-commerce application is developed using web technologies such as HTML, CSS, JavaScript, and Node.js, and is packaged into a Docker container. The container image is stored in Amazon Elastic Container Registry (ECR) and deployed using AWS ECS with the Fargate launch type.
The implementation highlights the benefits of cloud-based container orchestration, including scalability, reduced infrastructure management, and improved application availability. The results show that the proposed system provides efficient deployment, simplified maintenance, and reliable access through public internet connectivity. This study emphasizes the importance of modern DevOps practices in building scalable web applications.
The implementation highlights the benefits of cloud-based container orchestration, including scalability, reduced infrastructure management, and improved application availability. The results show that the proposed system provides efficient deployment, simplified maintenance, and reliable access through public internet connectivity. This study emphasizes the importance of modern DevOps practices in building scalable web applications.
Abstract
Serverless IoT Application Using AWS IoT Core and Lambda for Real-Time Sensor Data Processing
Dhanush K, Mr. B. Ramesh Kumar
DOI: 10.17148/IJARCCE.2026.15356
Abstract: The Internet of Things (IoT) is one of the most significant technological advancements in modern computing, enabling physical devices to communicate, exchange data, and perform intelligent operations through the internet. As the number of connected devices continues to grow rapidly, there is an increasing need for scalable, reliable, and secure cloud infrastructure capable of handling large volumes of sensor-generated data efficiently. This project focuses on the design and implementation of a Serverless IoT Application using Amazon Web Services (AWS) cloud technologies. The proposed system demonstrates how IoT sensor data can be collected, processed, stored, and monitored using a fully managed cloud-based architecture. Temperature data is transmitted securely to the cloud using the MQTT protocol and received by AWS IoT Core. An IoT rule triggers an AWS Lambda function that processes, validates, and stores the data in Amazon DynamoDB. Amazon CloudWatch provides monitoring and logging capabilities. The serverless architecture offers automatic scaling, reduced infrastructure management, improved reliability, and cost efficiency through a pay-per- use model.
Keywords: AWS IoT Core, AWS Lambda, Serverless Computing, Amazon DynamoDB, Amazon CloudWatch, MQTT Protocol, IoT, Cloud Computing, Python.
Keywords: AWS IoT Core, AWS Lambda, Serverless Computing, Amazon DynamoDB, Amazon CloudWatch, MQTT Protocol, IoT, Cloud Computing, Python.
Abstract
Kubernetes on aws with eks
Rijo, Dr. S. Thavamani
DOI: 10.17148/IJARCCE.2026.15357
Abstract: The rapid growth of cloud computing and microservice-based application development has increased the demand for efficient container orchestration platforms. Kubernetes has emerged as the leading solution for managing containerized workloads, offering automation, scalability, and high availability. At the same time, cloud providers such as Amazon Web Services (AWS) provide managed services that simplify Kubernetes cluster management.
This paper presents a practical implementation of deploying a containerized e-commerce web application using Kubernetes on Amazon Elastic Kubernetes Service (EKS). The system demonstrates how containerization with Docker, combined with Kubernetes orchestration and AWS cloud infrastructure, can create a scalable and reliable application deployment environment.
The architecture integrates multiple AWS services including Amazon EC2, Amazon VPC, Elastic Load Balancing, and Amazon CloudWatch to manage compute resources, networking, traffic distribution, and monitoring. The proposed solution improves application reliability, simplifies infrastructure management, and enables automatic scaling based on demand. Experimental deployment results show that Kubernetes on AWS EKS provides a robust platform for modern cloud-native applications.
This paper presents a practical implementation of deploying a containerized e-commerce web application using Kubernetes on Amazon Elastic Kubernetes Service (EKS). The system demonstrates how containerization with Docker, combined with Kubernetes orchestration and AWS cloud infrastructure, can create a scalable and reliable application deployment environment.
The architecture integrates multiple AWS services including Amazon EC2, Amazon VPC, Elastic Load Balancing, and Amazon CloudWatch to manage compute resources, networking, traffic distribution, and monitoring. The proposed solution improves application reliability, simplifies infrastructure management, and enables automatic scaling based on demand. Experimental deployment results show that Kubernetes on AWS EKS provides a robust platform for modern cloud-native applications.
Abstract
Build a Python Flask App and Deploy Using AWS Elastic Beanstalk
Vishaal S, Mr. S.S. SaravanaKumar
DOI: 10.17148/IJARCCE.2026.15358
Abstract: Web applications have become an essential part of modern digital services, enabling users to access systems and information from anywhere through the internet. With the increasing demand for scalable and efficient applications, cloud computing platforms provide reliable environments for deploying and managing web applications. This project focuses on the development and deployment of a web application using the Python Flask framework and Amazon Web Services (AWS) Elastic Beanstalk. The application is developed using Flask due to its lightweight structure, flexibility, and ease of integration with various web technologies. After development, the application is deployed on AWS Elastic Beanstalk, which simplifies the process of managing infrastructure, scaling resources, and maintaining application availability. The system demonstrates how cloud-based deployment can reduce configuration complexity while improving performance and scalability. Through this project, users can understand the complete workflow of designing, developing, and deploying a web application in a cloud environment. The results show that AWS Elastic Beanstalk provides an efficient platform for hosting Flask applications with minimal manual configuration. This project highlights the practical implementation of cloud deployment techniques and demonstrates the advantages of using modern cloud services for web application hosting and management.
Keywords: Python Flask, Web Application Development, Cloud Computing, AWS Elastic Beanstalk, Cloud Deployment, Scalable Web Applications.
Keywords: Python Flask, Web Application Development, Cloud Computing, AWS Elastic Beanstalk, Cloud Deployment, Scalable Web Applications.
Abstract
Multi-Cloud Deployment using Docker and Terraform
G. Ranganayaki, Dr. B. Narasimhan
DOI: 10.17148/IJARCCE.2026.15359
Abstract: Cloud computing has transformed the way applications are developed, deployed, and managed. Traditionally, organizations relied on a single cloud provider to host their applications and services. However, this approach introduces several challenges such as vendor lock-in, limited flexibility, higher risk of downtime, and dependency on a single infrastructure provider. To overcome these challenges, organizations are increasingly adopting a multi-cloud strategy, where applications and workloads are deployed across multiple cloud platforms. Multi-cloud deployment improves reliability, scalability, and performance while reducing dependency on a single cloud vendor. However, managing infrastructure across multiple cloud providers manually can be complex and time-consuming. This complexity arises due to differences in cloud services, configuration methods, and deployment processes. To address these challenges, automation tools such as Docker and Terraform are widely used in modern DevOps environments. Docker enables developers to package applications into lightweight containers that include all necessary dependencies, ensuring consistent behavior across different environments. Terraform, on the other hand, is an Infrastructure as Code (IaC) tool that allows developers to define and manage cloud infrastructure through configuration files. This project focuses on implementing a multi-cloud deployment architecture using Docker and Terraform. The system demonstrates how an application can be containerized using Docker and deployed across multiple cloud providers using Terraform scripts. Terraform automates the provisioning of infrastructure resources such as virtual machines, networking components, and security groups, while Docker ensures consistent application environments. The proposed system improves deployment efficiency, scalability, and reliability. It also reduces manual configuration errors and simplifies infrastructure management. By combining containerization and infrastructure automation, the project demonstrates a modern approach to deploying applications across multiple cloud platforms.
Keywords: Multi-Cloud Deployment, Docker, Terraform, DevOps, Infrastructure as Code, Containerization, Cloud Computing, AWS.
Keywords: Multi-Cloud Deployment, Docker, Terraform, DevOps, Infrastructure as Code, Containerization, Cloud Computing, AWS.
Abstract
AWS Step Functions for Orchestrating Serverless Workflows
Gijith M, Praveena A
DOI: 10.17148/IJARCCE.2026.15360
Abstract: In the modern digital era, managing student records manually through traditional standalone applications introduces challenges such as inconsistent validation, delayed notifications, and high infrastructure overhead. This project presents the design and implementation of a Serverless Student Record Management and Notification System using Amazon Web Services (AWS). The system leverages AWS Step Functions as the central workflow orchestration service, coordinating multiple serverless components to automate data validation, storage, and notification in a sequential pipeline. Student information submitted in JSON format is validated using AWS Lambda, securely stored in Amazon DynamoDB, and confirmed to administrators via automated email notifications through Amazon SNS. System execution is monitored in real time using Amazon CloudWatch. The proposed architecture eliminates dedicated server infrastructure, enables automatic scalability, and reduces operational costs, demonstrating a practical cloud-native approach to modern educational data management.
Keywords: AWS Step Functions, Serverless Computing, AWS Lambda, Amazon DynamoDB, Amazon SNS, Cloud Computing, Student Record Management, Workflow Orchestration, Python.
Keywords: AWS Step Functions, Serverless Computing, AWS Lambda, Amazon DynamoDB, Amazon SNS, Cloud Computing, Student Record Management, Workflow Orchestration, Python.
Abstract
AWS LAMBDA FOR SERVERLESS MICROSERVICES ARCHITECTURE
Ponnuguru M, Mr. T. Pradeep
DOI: 10.17148/IJARCCE.2026.15361
Abstract: E-commerce platforms require scalable and reliable backend infrastructures to support large volumes of user requests, transactions, and product management operations. Traditional monolithic architectures often face difficulties in handling dynamic workloads and require continuous server management. This research presents a serverless microservices-based ecommerce backend system using Amazon Web Services (AWS).
The proposed platform uses AWS Lambda for serverless computing, Amazon API Gateway for API management, DynamoDB for data storage, and messaging services such as SNS and SQS for asynchronous communication. The architecture divides the application into independent microservices including user management, product catalog, cart management, order processing, and payment services.
Infrastructure deployment is automated using AWS Serverless Application Model (SAM). The serverless approach reduces infrastructure overhead, improves scalability, and enables costefficient cloud resource utilization. The system demonstrates how cloud-native services can be integrated to build a modern, scalable, and reliable e-commerce backend platform.
Keywords: AWS Lambda, Serverless Computing, Microservices, E-commerce, DynamoDB, Cloud Computing
The proposed platform uses AWS Lambda for serverless computing, Amazon API Gateway for API management, DynamoDB for data storage, and messaging services such as SNS and SQS for asynchronous communication. The architecture divides the application into independent microservices including user management, product catalog, cart management, order processing, and payment services.
Infrastructure deployment is automated using AWS Serverless Application Model (SAM). The serverless approach reduces infrastructure overhead, improves scalability, and enables costefficient cloud resource utilization. The system demonstrates how cloud-native services can be integrated to build a modern, scalable, and reliable e-commerce backend platform.
Keywords: AWS Lambda, Serverless Computing, Microservices, E-commerce, DynamoDB, Cloud Computing
Abstract
Smart IoT-Based Stress Detection & Autonomous Relaxation System
Prof. V Chaudhari, Devanshu A. Bidkar, Tanvi V. Bodhe, Ansh K. Kapse, Ruchi S. Rajput, Ujwal S. Sewaiwar
DOI: 10.17148/IJARCCE.2026.15362
Abstract: In contemporary high-pressure corporate and academic environments, physiological stress has emerged as a significant yet often undetected precursor to chronic health conditions and diminished productivity. Current wellness interventions are predominantly reactive, requiring users to manually initiate relaxation only after the onset of exhaustion. This delay often results in long-term health complications and a lack of sustainable mental well-being in the workspace.
The objective of this proposed system is to develop an "IoT-Enabled Autonomous Stress Detection and Relaxation Chair" that provides proactive, non-invasive health support. The system utilizes an ESP32-based architecture interfaced with a multi-sensor array, including biomedical sensors for heart rate monitoring and thermal sensors for body temperature tracking. By integrating these components into a standard seating environment, the chair acts as a continuous health- monitoring assistant.
The methodology follows a "Sense-Analyze-Act-Monitor" feedback loop. Physiological data is continuously polled and processed through embedded algorithms to classify stress states. Upon identifying a stress spike, the system autonomously triggers an integrated relaxation cycle consisting of targeted vibration massage, controlled heat therapy, and calming auditory stimulation. Simultaneously, real-time data is synchronized with a cloud-based dashboard using the HiveMQTT broker for long-term health visualization.
The objective of this proposed system is to develop an "IoT-Enabled Autonomous Stress Detection and Relaxation Chair" that provides proactive, non-invasive health support. The system utilizes an ESP32-based architecture interfaced with a multi-sensor array, including biomedical sensors for heart rate monitoring and thermal sensors for body temperature tracking. By integrating these components into a standard seating environment, the chair acts as a continuous health- monitoring assistant.
The methodology follows a "Sense-Analyze-Act-Monitor" feedback loop. Physiological data is continuously polled and processed through embedded algorithms to classify stress states. Upon identifying a stress spike, the system autonomously triggers an integrated relaxation cycle consisting of targeted vibration massage, controlled heat therapy, and calming auditory stimulation. Simultaneously, real-time data is synchronized with a cloud-based dashboard using the HiveMQTT broker for long-term health visualization.
Abstract
AI COMPLAINT-PRIORITY SYSTEM FOR LOCAL GOVERNMENT
Dr. B. Sivaranjani, Mr. B. Dharneesh, Mr. S. Dharaneesh, Mr. K. Dharshan, Mrs. K. Abirami
DOI: 10.17148/IJARCCE.2026.15363
Abstract: The AI Complaint-Priority System for local government is a smart and easy-to-use platform where citizens can report problems like potholes, water issues, and noise complaints through a mobile app or website. Python is used to read and understand the complaints, while MySQL safely stores all user and complaint data.The AI automatically checks the urgency of each complaint by analyzing the words, location, and seriousness, then sends it to the correct department. A clean dashboard made with CSS shows real-time updates, reduces delays, avoids manual mistakes, and helps the government fix issues faster. It also identifies repeated problems in an area, improving planning and building trust with citizens.The system also provides automatic notifications to inform citizens about the status of their complaints. It keeps a complete history of complaints for future reference and analysis. Government officials can monitor complaint resolution progress through the dashboard. The platform supports data analytics and reports to help authorities understand common issues in different areas. This system improves transparency, accountability, and communication between citizens and local government authorities, making public service management more efficient.
Keywords: Artificial intelligence Complaint management System, Natural Language Processing Python, MySQL Database, Automated Complaint Prioritization.
Keywords: Artificial intelligence Complaint management System, Natural Language Processing Python, MySQL Database, Automated Complaint Prioritization.
Abstract
Deploy a Static Website on AWS S3 and Use CloudFront
Yaswanth S, Mr. B. Ramesh Kumar
DOI: 10.17148/IJARCCE.2026.15364
Abstract: Cloud computing has transformed the way websites are hosted and delivered to users around the world. Traditional hosting methods often require complex infrastructure and maintenance. Amazon Web Services (AWS) provides scalable and cost-effective solutions to deploy websites quickly and efficiently.
This project demonstrates how to deploy a static website using Amazon S3 and improve its performance and security using Amazon CloudFront. Amazon S3 is used to store static website files such as HTML, CSS, and JavaScript, while CloudFront acts as a Content Delivery Network (CDN) that distributes the website content globally with low latency.
The implementation includes creating an S3 bucket, uploading website files, enabling static website hosting, configuring permissions, and integrating CloudFront for faster content delivery. The project highlights how cloud services simplify website deployment and enhance performance.
This project demonstrates how to deploy a static website using Amazon S3 and improve its performance and security using Amazon CloudFront. Amazon S3 is used to store static website files such as HTML, CSS, and JavaScript, while CloudFront acts as a Content Delivery Network (CDN) that distributes the website content globally with low latency.
The implementation includes creating an S3 bucket, uploading website files, enabling static website hosting, configuring permissions, and integrating CloudFront for faster content delivery. The project highlights how cloud services simplify website deployment and enhance performance.
Abstract
NeuroSpill: Attention U-Net++ Based Oil Spill Segmentation in SAR Imagery
Aishmi Anish, Ance Koshy, Nebu Placid, Sunu Sara Antony, Neethu Thomas
DOI: 10.17148/IJARCCE.2026.15365
Abstract: Marine oil spills pose a significant threat to ocean ecosystems, marine biodiversity, and coastal environments. Rapid detection and monitoring of oil spills are essential for minimising environmental damage and enabling effective response strategies. Synthetic Aperture Radar (SAR) imagery has emerged as a reliable remote sensing tool for oil spill detection due to its ability to capture ocean surface information under all weather and illumination conditions. However, accurately distinguishing oil spills from look-alike phenomena in SAR images remains challenging due to speckle noise, complex ocean dynamics, and limited annotated datasets. This paper presents NeuroSpill, a deep learning based framework for automated oil-spill detection using SAR imagery. The proposed method employs an Attention-UNet++ architecture to perform precise semantic segmentation of oil spill regions. To enhance feature representation, three SAR- derived channels—VV polarisation, VH polarisation, and the VV/VH ratio—are utilised as input features. A patch-wise prediction strategy is implemented to process large SAR scenes while improving segmentation accuracy efficiently. In addition, a verification mechanism and morphological post-processing are applied to reduce noise and refine spill boundaries. The framework also incorporates an oil spill analysis module that estimates spill coverage and categorises severity levels. An interactive Streamlit-based interface is developed to enable real-time visualisation and monitoring of SAR images. Experimental results demonstrate that the proposed system provides reliable oil spill detection and offers a practical solution for automated marine environmental monitoring.
Keywords: Attention U-Net++, Deep Learning, Environmental Monitoring, Oil Spill Detection, Remote Sensing, Synthetic Aperture Radar, SAR Semantic Segmentation
Keywords: Attention U-Net++, Deep Learning, Environmental Monitoring, Oil Spill Detection, Remote Sensing, Synthetic Aperture Radar, SAR Semantic Segmentation
Abstract
The Tech-Integrated Charity Chain for Transparent Giving and Efficient Aid Delivery
Vinay S. Khapare, Pranav D. Khot, Arkaj U. Kadam, Shreyash P.Alase, Jaydeep A. Patil, Prof. Sambhaji D. Rane
DOI: 10.17148/IJARCCE.2026.15366
Abstract: Charitable organizations play a significant role in addressing social challenges such as poverty, disaster relief, healthcare support, and educational assistance. However, traditional donation systems often lack transparency, accountability, and efficient fund tracking, which can reduce donor trust and participation. This research proposes The Tech-Integrated Charity Chain, a digital platform designed to improve transparency and efficiency in charitable donations through real-time tracking, verified NGO management, and impact visualization.
The system utilizes modern web technologies to provide a secure and scalable framework for managing donors, NGOs, campaigns, and beneficiaries. By enabling real-time monitoring of donations and campaign progress, the proposed platform enhances donor confidence and ensures responsible fund utilization. Experimental evaluation demonstrates improved transparency, faster reporting, and better user engagement compared to conventional charity systems.
Keywords: Charity Platform, Donation Transparency, NGO Verification, Real-Time Tracking, Web Application, Aid Management.
The system utilizes modern web technologies to provide a secure and scalable framework for managing donors, NGOs, campaigns, and beneficiaries. By enabling real-time monitoring of donations and campaign progress, the proposed platform enhances donor confidence and ensures responsible fund utilization. Experimental evaluation demonstrates improved transparency, faster reporting, and better user engagement compared to conventional charity systems.
Keywords: Charity Platform, Donation Transparency, NGO Verification, Real-Time Tracking, Web Application, Aid Management.
Abstract
DEPLOY A MULTI-REGION APPLICATION
WITH DOCKER AND AWS ROUTE, Pon Vignesh P, Gowri Lakshmi K S
DOI: 10.17148/IJARCCE.2026.15367
Abstract: The modern digital economy demands "always-on" application availability, where even brief downtime can result in significant financial loss and eroded consumer trust. Deploying a multi-region application using Docker and AWS Route 53 represents a transition from traditional high availability to a "global-first" architecture. While standard cloud deployments often rely on multiple Availability Zones within a single geographic region, a truly resilient system must account for regional-level failures such as large-scale power outages, natural disasters, or fiber optic interruptions. Docker enables organizations to package microservices into immutable, portable units that function identically across regions — whether hosted in Virginia, Dublin, or Sydney — eliminating "environmental drift" that historically plagued cross-site disaster recovery.
The technical foundation rests on deploying Amazon Elastic Container Service (ECS) or Amazon Elastic Kubernetes Service (EKS) clusters across geographically distinct AWS regions. Each region acts as a fully functional, autonomous silo with its own VPC, subnets, and Application Load Balancers (ALBs). Docker images are stored in Amazon Elastic Container Registry (ECR) with cross-region replication, ensuring low deployment latency. AWS Route 53 manages DNS and intelligently routes users to the nearest healthy region, providing sub-second response times and high fault tolerance.
Keywords: Docker, AWS Route 53, Multi-Region Deployment, Cloud Computing, High Availability, Containerization, DNS Routing, Fault Tolerance
The technical foundation rests on deploying Amazon Elastic Container Service (ECS) or Amazon Elastic Kubernetes Service (EKS) clusters across geographically distinct AWS regions. Each region acts as a fully functional, autonomous silo with its own VPC, subnets, and Application Load Balancers (ALBs). Docker images are stored in Amazon Elastic Container Registry (ECR) with cross-region replication, ensuring low deployment latency. AWS Route 53 manages DNS and intelligently routes users to the nearest healthy region, providing sub-second response times and high fault tolerance.
Keywords: Docker, AWS Route 53, Multi-Region Deployment, Cloud Computing, High Availability, Containerization, DNS Routing, Fault Tolerance
Abstract
School Management System
Lect. Kalyani Wankhede, Arpit Raut, Gaurav Kamble, Ashutosh Wankhede, Shubham Somkuwar, Khushi Gamey, Dimple Madavi.
DOI: 10.17148/IJARCCE.2026.15368
Abstract: The management of student academic records in educational institutions is traditionally performed through manual record-keeping methods such as registers, spreadsheets, and paper documents. These approaches are often time- consuming, prone to human errors, and inefficient when retrieving or updating information. To overcome these limitations, a web-based School Management System has been developed to digitize and streamline the management of student information.
The system allows parents to securely log in and view important details such as student attendance and subject-wise academic marks. Teachers are provided with an administrative dashboard through which they can manage student data, update attendance records, and upload examination marks.
The proposed system is implemented using Angular for the frontend interface, C# for backend processing, and a structured database for storing student records. By integrating authentication mechanisms and centralized data management, the system ensures secure and reliable access to academic information. The automated platform improves transparency, reduces administrative workload, and enhances communication between teachers and parents.
Keywords: School Management System, Student Portal, Angular, C#, Web Application, Attendance Management, Academic Records.
The system allows parents to securely log in and view important details such as student attendance and subject-wise academic marks. Teachers are provided with an administrative dashboard through which they can manage student data, update attendance records, and upload examination marks.
The proposed system is implemented using Angular for the frontend interface, C# for backend processing, and a structured database for storing student records. By integrating authentication mechanisms and centralized data management, the system ensures secure and reliable access to academic information. The automated platform improves transparency, reduces administrative workload, and enhances communication between teachers and parents.
Keywords: School Management System, Student Portal, Angular, C#, Web Application, Attendance Management, Academic Records.
Abstract
SERVERLESS IMAGE RECOGNITION WITH AWS REKOGNITION AND LAMBDA
Menaka Sri S, Dr. Shylaja M
DOI: 10.17148/IJARCCE.2026.15369
Abstract: The rapid proliferation of visual data in the digital era has necessitated advanced, scalable, and cost-effective solutions for automated image analysis. Traditional server-bound architectures often struggle with the unpredictable nature of image processing workloads, leading to either resource underutilization or high latency during peak traffic. This paper presents an end-to-end serverless architecture designed to perform real-time image recognition and metadata extraction using Amazon Rekognition and AWS Lambda. By leveraging the Function-as-a-Service (FaaS) model, the system eliminates the need for manual server provisioning, patching, and scaling, providing a robust framework that responds dynamically to incoming data streams.
At the core of this system is an event-driven workflow initiated by the ingestion of visual assets into Amazon S3. The upload event triggers an AWS Lambda function that invokes the Amazon Rekognition API — a deep-learning-based service capable of identifying objects, scenes, text, and facial attributes without requiring specialized machine learning expertise. Results comprising detected labels and confidence scores are persisted in Amazon DynamoDB, enabling massive parallelization that can handle thousands of simultaneous uploads with minimal latency. The system also supports facial analysis, celebrity recognition, and content moderation using a pay-as-you-go pricing model.
Keywords: Serverless Computing, AWS Lambda, Amazon Rekognition, Image Recognition, Amazon S3, DynamoDB, Cloud Computing, FaaS, Event-Driven Architecture, Deep Learning
At the core of this system is an event-driven workflow initiated by the ingestion of visual assets into Amazon S3. The upload event triggers an AWS Lambda function that invokes the Amazon Rekognition API — a deep-learning-based service capable of identifying objects, scenes, text, and facial attributes without requiring specialized machine learning expertise. Results comprising detected labels and confidence scores are persisted in Amazon DynamoDB, enabling massive parallelization that can handle thousands of simultaneous uploads with minimal latency. The system also supports facial analysis, celebrity recognition, and content moderation using a pay-as-you-go pricing model.
Keywords: Serverless Computing, AWS Lambda, Amazon Rekognition, Image Recognition, Amazon S3, DynamoDB, Cloud Computing, FaaS, Event-Driven Architecture, Deep Learning
Abstract
Smart Speak Home Automation
Prof. Shital Hajare, Jagdish A. Neware, Jivesh A. Karnuke, Nishant D. Mahajan, Ganesh S. Parwe, Ankita L. Barapatre
DOI: 10.17148/IJARCCE.2026.15370
Abstract: The development of Internet of Things (IoT) technology has significantly improved smart home automation systems. However, many existing systems rely on cloud connectivity which may raise concerns related to privacy and internet dependency. This paper presents a Smart Speak Home Automation System using the VC-02 offline voice recognition module. The system processes voice commands locally without internet connectivity and controls electrical appliances using relay modules connected to an Arduino UNO microcontroller. In addition, safety monitoring is implemented using MQ-2 gas sensors and to detect hazardous conditions. Wireless communication between modules is achieved using RF transmitter and receiver modules with HT12E encoder. The system provides reliable appliance control, fast response time and improved user privacy.
Keywords: IoT, Smart Home Automation, VC-02 Voice Module, Arduino UNO, RF Communication, MQ-2 Gas Sensor.
Keywords: IoT, Smart Home Automation, VC-02 Voice Module, Arduino UNO, RF Communication, MQ-2 Gas Sensor.
Abstract
Secure Web Application Using Docker, Nginx, and AWS WAF
Jagatheeshwaran E, Dr. C. Daniel Nesakumar
DOI: 10.17148/IJARCCE.2026.15371
Abstract: The rapid expansion of internet-based services has increased the demand for secure, scalable, and reliable web application deployment environments. Traditional deployment methods often face challenges such as inconsistent environments, security vulnerabilities, and difficulty managing application scalability. This project presents the design and implementation of a secure web application deployment environment using Docker, Nginx, and AWS Web Application Firewall (WAF). Docker provides containerization to ensure application portability and consistency across development, testing, and production environments. Nginx acts as a reverse proxy server handling request routing, load balancing, and security filtering. AWS WAF monitors and blocks malicious HTTP/HTTPS traffic including SQL injection and cross-site scripting (XSS) attacks. The infrastructure is hosted on Amazon EC2 instances, providing scalability and high availability. The proposed system improves security, simplifies deployment, and ensures consistent performance across different environments.
Keywords: Docker, Containerization, Nginx, Reverse Proxy, AWS WAF, Web Security, Amazon EC2, Cloud Computing, SQL Injection, XSS Protection.
Keywords: Docker, Containerization, Nginx, Reverse Proxy, AWS WAF, Web Security, Amazon EC2, Cloud Computing, SQL Injection, XSS Protection.
Abstract
CNN-Based Dorsal Hand Vein Authentication Using Triplet Loss Metric Learning
J Adithye, Abhimanyu R, Abhijith S Kurup, Adithiya S, Girija V R
DOI: 10.17148/IJARCCE.2026.15372
Abstract: Hand vein biometrics has emerged as a reliable authentication modality due to the uniqueness and internal nature of vascular patterns. Unlike surface-based biometrics such as fingerprints or facial recognition, vein patterns are less susceptible to environmental variations and spoofing attacks. This work presents a CNN-based dorsal hand vein biometric authentication framework that combines a ten-stage preprocessing pipeline with deep metric learning-based feature extraction. The preprocessing pipeline includes Gaussian smoothing, automatic hand region detection, wrist removal using distance-transform thickness profiling, finger removal using convex hull-based techniques, Otsu segmentation, CLAHE contrast enhancement, Sato vesselness filtering, feathered mask application, and tight crop resize to 224Ă—224 pixels to isolate the stable metacarpal vein region.
A four-block CNN backbone is trained using online semi-hard triplet loss mining with an identity-based batch sampler to generate discriminative 128-dimensional L2-normalized embeddings for biometric matching. A manually collected dataset of 5220 dorsal hand vein images from 261 individuals (522 hand identities, with left and right hands treated as separate biometric identities) captured under controlled near-infrared illumination was used for training and evaluation. Experimental results demonstrate a recognition accuracy of 99.12% and an Equal Error Rate of 1.32%, confirming the effectiveness of the proposed framework for secure hand vein biometric authentication.
Keywords: Authentication, Biometric Recognition, Convolutional Neural Network (CNN), Dorsal Hand Vein Recognition, Embedding, Metric Learning, Near-Infrared Imaging, Online Semi-Hard Mining, Triplet Loss, Vessel Enhancement.
A four-block CNN backbone is trained using online semi-hard triplet loss mining with an identity-based batch sampler to generate discriminative 128-dimensional L2-normalized embeddings for biometric matching. A manually collected dataset of 5220 dorsal hand vein images from 261 individuals (522 hand identities, with left and right hands treated as separate biometric identities) captured under controlled near-infrared illumination was used for training and evaluation. Experimental results demonstrate a recognition accuracy of 99.12% and an Equal Error Rate of 1.32%, confirming the effectiveness of the proposed framework for secure hand vein biometric authentication.
Keywords: Authentication, Biometric Recognition, Convolutional Neural Network (CNN), Dorsal Hand Vein Recognition, Embedding, Metric Learning, Near-Infrared Imaging, Online Semi-Hard Mining, Triplet Loss, Vessel Enhancement.
Abstract
AI-Based Vegetable Disease Detection
Mr. H.M. Gaikwad, Ms. V. N. Lawand, Tanuja Ahire, Shrushti Dhamode
DOI: 10.17148/IJARCCE.2026.15373
Abstract: Plant diseases in vegetable crops significantly affect agricultural productivity and yield. Traditional disease detection methods depend on manual inspection, which is time-consuming and prone to errors. To address this issue, this project uses Convolutional Neural Networks (CNNs) to automatically detect vegetable leaf diseases through image classification.The system classifies vegetable leaves into healthy and diseased categories, including bacterial and fungal infections, blight, and leaf spot diseases. Using a labeled dataset and image preprocessing techniques such as resizing and normalization, the CNN model learns disease-specific features effectively. This AI-based system enables early disease detection, reduces crop loss, improves yield, and supports sustainable farming practices, with future scope for mobile deployment and multi-crop expansion.
Keywords: Vegetable Disease Detection, CNN, Deep Learning, Image Classification, Smart Agriculture
Keywords: Vegetable Disease Detection, CNN, Deep Learning, Image Classification, Smart Agriculture
Abstract
Internet of Things (IoT): Architecture, Applications, Challenges, and Future Scope
AKHIL.V.R, SANTHOSH.V, Dr. P. RADHA
DOI: 10.17148/IJARCCE.2026.15374
Abstract: The Internet of Things (IoT) has emerged as one of the most transformative technological paradigms of the 21st century, fundamentally redefining the way physical objects interact with the digital world. By embedding sensors, actuators, and communication modules into everyday devices, IoT enables seamless data collection, real-time monitoring, and intelligent automation across a wide spectrum of industries. This paper provides a comprehensive review of IoT architecture, encompassing the perception, network, and application layers. It examines widely deployed communication protocols including MQTT, CoAP, HTTP/HTTPS, Zigbee, and Z-Wave, and discusses cloud and edge computing platforms that underpin modern IoT deployments. Key application domains are explored in depth, including smart homes, smart cities, industrial IoT (IIoT), precision agriculture, healthcare monitoring, and connected transportation. The paper further analyzes critical challenges such as heterogeneity of devices, interoperability, security vulnerabilities, scalability bottlenecks, energy constraints, and data privacy concerns. Finally, emerging trends including the convergence of IoT with Artificial Intelligence (AI), 5G networks, blockchain, and digital twins are discussed, along with their implications for future IoT ecosystems.
Keywords: Internet of Things, IoT Architecture, Smart Devices, Edge Computing, IoT Security, Industrial IoT, Smart Cities, Wireless Sensor Networks, Embedded Systems, Cloud Computing.
Keywords: Internet of Things, IoT Architecture, Smart Devices, Edge Computing, IoT Security, Industrial IoT, Smart Cities, Wireless Sensor Networks, Embedded Systems, Cloud Computing.
Abstract
Fruit Grade
Prof. Prof. Smita Thakare, Pallavi Jankiram Solanke, Bhagyashri Mahendra Pardeshi, Tanisha Ajay Lokhande, Parth Prabhanja Naik
DOI: 10.17148/IJARCCE.2026.15375
Abstract: Maintaining consistent fruit quality evaluation and establishing efficient market connectivity are major challenges in the modern agricultural supply chain. Traditional fruit grading methods rely heavily on manual inspection, which can lead to inconsistent results, human error, and inaccurate pricing. Such limitations often cause financial losses for farmers and vendors while also affecting the reliability of quality standards in the market.
The proposed SmartFruit — AI-powered Fruit Quality Analysis & Vendor Connect Platform introduces an intelligent solution that utilizes Artificial Intelligence (AI) and Machine Learning (ML) techniques to automate fruit quality assessment. The system analyzes fruit images captured through a mobile or web-based application and classifies them into predefined quality categories such as Grade A, Grade B, and Grade C. Using image processing and deep learning models, the platform evaluates visual attributes including color consistency, ripeness level, surface texture, and visible defects to determine the overall quality of fruits.
In addition to automated grading, the platform integrates a vendor connectivity module that helps farmers identify nearby buyers or vendors, compare potential offers, and maintain digital records of their transactions. The application also provides user-friendly features such as text-based results, optional voice feedback, and historical analysis tracking to improve accessibility and usability.
By combining automated fruit grading with digital market connectivity, the SmartFruit platform enhances transparency in quality evaluation, reduces dependence on manual inspection, and supports efficient decision-making within the agricultural ecosystem. The system ultimately contributes to minimizing post-harvest losses and strengthening the direct connection between farmers and market stakeholders.
Keywords: Artificial Intelligence (AI), Machine Learning, Fruit Quality Classification, Computer Vision, Image Processing, Smart Agriculture Systems, Vendor Connectivity Platform, AI-Based Fruit Grading, Agricultural Supply Chain Management.
The proposed SmartFruit — AI-powered Fruit Quality Analysis & Vendor Connect Platform introduces an intelligent solution that utilizes Artificial Intelligence (AI) and Machine Learning (ML) techniques to automate fruit quality assessment. The system analyzes fruit images captured through a mobile or web-based application and classifies them into predefined quality categories such as Grade A, Grade B, and Grade C. Using image processing and deep learning models, the platform evaluates visual attributes including color consistency, ripeness level, surface texture, and visible defects to determine the overall quality of fruits.
In addition to automated grading, the platform integrates a vendor connectivity module that helps farmers identify nearby buyers or vendors, compare potential offers, and maintain digital records of their transactions. The application also provides user-friendly features such as text-based results, optional voice feedback, and historical analysis tracking to improve accessibility and usability.
By combining automated fruit grading with digital market connectivity, the SmartFruit platform enhances transparency in quality evaluation, reduces dependence on manual inspection, and supports efficient decision-making within the agricultural ecosystem. The system ultimately contributes to minimizing post-harvest losses and strengthening the direct connection between farmers and market stakeholders.
Keywords: Artificial Intelligence (AI), Machine Learning, Fruit Quality Classification, Computer Vision, Image Processing, Smart Agriculture Systems, Vendor Connectivity Platform, AI-Based Fruit Grading, Agricultural Supply Chain Management.
Abstract
Insider Threat Detection using Agentic AI
R.Y. Thombare, S.V. Waghmare, Amey Malpurkar, Aditya Marathe, Chhatraraj Patil, Tanmay Gaikwad
DOI: 10.17148/IJARCCE.2026.15376
Abstract: Organizations today rely heavily on digital infrastructure to store and manage sensitive information such as financial records, intellectual property, and confidential documents. While traditional cybersecurity solutions effectively protect against external attacks, threats originating from authorized internal users remain difficult to detect. Insider threats occur when employees, contractors, or partners misuse their legitimate access privileges either intentionally or unintentionally. Conventional security systems mainly rely on rule-based monitoring and signature detection, which often fail to identify subtle behavioral anomalies.
This research proposes an intelligent insider threat detection framework using Agentic Artificial Intelligence. The system utilizes autonomous AI agents that continuously monitor user behavior across multiple data sources including system logs, network traffic, and file access records. Machine learning techniques are used to build behavioral profiles and detect deviations from normal patterns. The agentic architecture enables reasoning over anomalies and supports automated threat assessment.
Experimental evaluation using publicly available cybersecurity datasets demonstrates that the proposed approach improves threat detection accuracy and reduces false positives compared to traditional systems. The system provides real-time alerts and contextual explanations, enabling security teams to respond quickly to potential risks. This research highlights the potential of Agentic AI to enhance modern cybersecurity systems by enabling proactive detection and intelligent response to insider threats.
Keywords: Insider Threat Detection, Agentic AI, Cybersecurity, Behavioral Analysis, Machine Learning, Anomaly Detection, User Activity Monitoring.
This research proposes an intelligent insider threat detection framework using Agentic Artificial Intelligence. The system utilizes autonomous AI agents that continuously monitor user behavior across multiple data sources including system logs, network traffic, and file access records. Machine learning techniques are used to build behavioral profiles and detect deviations from normal patterns. The agentic architecture enables reasoning over anomalies and supports automated threat assessment.
Experimental evaluation using publicly available cybersecurity datasets demonstrates that the proposed approach improves threat detection accuracy and reduces false positives compared to traditional systems. The system provides real-time alerts and contextual explanations, enabling security teams to respond quickly to potential risks. This research highlights the potential of Agentic AI to enhance modern cybersecurity systems by enabling proactive detection and intelligent response to insider threats.
Keywords: Insider Threat Detection, Agentic AI, Cybersecurity, Behavioral Analysis, Machine Learning, Anomaly Detection, User Activity Monitoring.
Abstract
Deepfake Audio Detection Using Hybrid Random Forest and Convolutional Neural Network Architecture
K Dharma Ratnam, P Kanaka Tulasi
DOI: 10.17148/IJARCCE.2026.15377
Abstract: The rapid evolution of speech synthesis and voice conversion technologies has enabled the generation of highly realistic synthetic speech, commonly referred to as deepfake audio. While such technologies offer innovative applications in media and accessibility, they also introduce serious threats to security, privacy, and information authenticity. This paper presents a hybrid deepfake audio detection system that combines classical machine learning and deep learning techniques to identify spoofed speech. The proposed framework integrates a Random Forest classifier trained on Mel- Frequency Cepstral Coefficients (MFCCs) and a Convolutional Neural Network (CNN) trained on Log-Mel Spectrogram representations. The system is implemented as a standalone desktop application with real-time visualization support. Experimental results demonstrate that the hybrid approach achieves high classification accuracy while maintaining computational efficiency suitable for consumer-grade hardware. The proposed solution aims to provide an accessible and reliable tool for combating synthetic audio misuse.
Keywords: Deepfake Audio, Audio Spoofing Detection, MFCC, CNN, Random Forest, Spectrogram Analysis, Anti- Spoofing
Keywords: Deepfake Audio, Audio Spoofing Detection, MFCC, CNN, Random Forest, Spectrogram Analysis, Anti- Spoofing
Abstract
An AI-Powered Interior, Exterior Designer & Layout Planner
HOD R. Y. Thombare, Mr. S. V. Waghmare, Sarthak Deshpande, Atharva Ahire, Aditya Awasarkar, Mitesh Pawar
DOI: 10.17148/IJARCCE.2026.15378
Abstract: Interior and exterior design planning plays an important role in residential and commercial construction, but traditional methods require significant time, cost, and professional expertise. To address these challenges, this project proposes an AI-powered Interior, Exterior, and Layout Planner Designing Website that helps users create smart, aesthetic, and optimized design solutions.
The system uses Artificial Intelligence and Machine Learning techniques to generate interior layouts, exterior designs, and space planning recommendations based on user inputs such as room size, building type, style preferences, furniture needs, and budget. It provides multiple design suggestions and an interactive platform to visualize layouts, furniture placement, and color combinations.
This solution reduces manual effort, improves personalization, and offers a cost-effective and userfriendly approach to modern architectural design, with future scope for 3D visualization and AR/VR integration.
Keywords: Artificial Intelligence, Interior Design, Exterior Design, Layout Planning, Machine Learning, Smart Architecture, Web Application.
The system uses Artificial Intelligence and Machine Learning techniques to generate interior layouts, exterior designs, and space planning recommendations based on user inputs such as room size, building type, style preferences, furniture needs, and budget. It provides multiple design suggestions and an interactive platform to visualize layouts, furniture placement, and color combinations.
This solution reduces manual effort, improves personalization, and offers a cost-effective and userfriendly approach to modern architectural design, with future scope for 3D visualization and AR/VR integration.
Keywords: Artificial Intelligence, Interior Design, Exterior Design, Layout Planning, Machine Learning, Smart Architecture, Web Application.
Abstract
Skin Guard Pro: AI Powered Skin Disease Detection System.
Mr. H.M. Gaikwad, Vansha Nikam, Sawant Poorva, More Lokesh, Yeola Vaishnavi
DOI: 10.17148/IJARCCE.2026.15379
Abstract: Skin diseases require early detection, continuous monitoring, and timely medical intervention, yet traditional healthcare systems often lack real-time screening solutions, structured digital records, and accessible dermatological support beyond hospital visits. Challenges such as delayed consultations, misdiagnosis, poor awareness of symptoms, irregular follow-ups, limited availability of specialists in rural areas, and subjective visual examination reduce treatment effectiveness and may lead to serious complications. To address these issues, this project proposes Skin Guard Pro – an AI Powered Skin Disease Detection System that integrates intelligent image analysis, symptom-based evaluation, real- time disease classification, preventive guidance, and secure data management within a unified digital platform. The system enables users to upload skin images, enter symptoms, and receive AI-driven predictions using advanced deep learning techniques such as Convolutional Neural Networks (CNNs). It further provides structured disease information, precautionary measures, and recommendations for medical consultation when necessary. By digitizing preliminary dermatological screening and ensuring accessible, data-driven diagnosis support, the proposed solution enhances early detection, improves patient awareness, promotes proactive care, and enables more efficient, reliable, and personalized skin healthcare management.
Keywords: Artificial Intelligence (AI), Skin Disease Detection, Convolutional Neural Network (CNN), Deep Learning, Medical Image Processing, Dermatology, Image Classification, Early Diagnosis, Healthcare Technology, Digital Health Systeme
Keywords: Artificial Intelligence (AI), Skin Disease Detection, Convolutional Neural Network (CNN), Deep Learning, Medical Image Processing, Dermatology, Image Classification, Early Diagnosis, Healthcare Technology, Digital Health Systeme
Abstract
“AI ASSISTED PERSONAL ROBOT”
Prof. Bhagyashree Wankhede, Aarju S. Dhurve, Lokesh P. Kothe, Mitali S. Mankar, Pranjali A. Nagose, Swejal D. Patil
DOI: 10.17148/IJARCCE.2026.15380
Abstract: The rapid growth of intelligent personal assistants has largely been driven by cloud-based artificial intelligence services, which inherently suffer from privacy concerns, network dependency, and non-deterministic latency. These limitations become critical in mobile robotic platforms where timely and reliable responses are essential for safe navigation and natural interaction. This paper presents the design and implementation of an AI-assisted personal robot that integrates a locally hosted Large Language Model (LLM), specifically Llama 3, with an ESP32-S3 microcontroller performing edge computing. The proposed system employs an I2S-based digital audio acquisition pipeline using an INMP441 microphone, a WebSocket-based bidirectional communication framework between the robot and a local AI server, and an ultrasonic-based safety interrupt mechanism for collision avoidance. Speech is captured as raw PCM data on the ESP32-S3, transmitted over Wi-Fi to a local workstation running Whisper for speech-to-text conversion and Llama 3 for intent understanding, and returned as structured JSON commands for real- time motor control. Experimental evaluation demonstrates sub-second end-to-end latency, robust command recognition in the presence of motor noise, and reliable indoor navigation under constrained network conditions. The results indicate that local LLM-driven robotics on commodity hardware can achieve low-latency, privacy-preserving interaction without dependence on cloud infrastructure, providing a viable architectural template for future embodied AI systems
Keywords: ESP32-S3, Edge AI, Local LLM, I2S Protocol, Web Socket Communication, Privacy-Preserving Robotics
Keywords: ESP32-S3, Edge AI, Local LLM, I2S Protocol, Web Socket Communication, Privacy-Preserving Robotics
Abstract
3D HUMANOID AI-SMART "AW8R”
Prof. CHANDRESH BHANGE, PRATYUSH.G. JANBHANDHU, SAMIKSHA A. GAJBHIYE, PRAKRUTI R. GAIMUKHE, TANU R. PATALE, RAVISHANKAR V. SHAHU, JAYASHRI M. MARBATE
DOI: 10.17148/IJARCCE.2026.15381
Abstract: The 3D Humanoid Al-Smart “AW8R” is an advanced portable Al-based voice assistant system designed to enhance natural human-machine interaction by integrating speech recognition, emotion detection, and a 3D holographic humanoid avatar. Unlike traditional voice assistants that rely solely on audio responses, the proposed system introduces synchronized facial expressions and lip movements to create a more engaging and realistic user experience.
The system is built around the ESP32-S3 microcontroller with integrated Wi-Fi for cloud connectivity. Audio input is captured using an INMP441 MEMS microphone and transmitted to cloud-based Al services such as Gemini or Open Al for speech-to-text conversion, intent detection, and contextual response generation. The backend also performs emotion analysis and text-to-speech synthesis. The processed response is delivered to the user through a 3W speaker driven by a MAX98357A amplifier, while a 1.3-inch TFT display renders a 3D humanoid avatar. A holographic acrylic pyramid creates a floating 3D visual effect.
The system is powered by a 2400mAh Li-ion battery with integrated charging and boost regulation. Experimental evaluation demonstrates stable operation, synchronized animation, and improved user engagement compared to traditional voice-only systems. The project provides a cost-effective and scalable solution for next-generation Al assistants.
The system is built around the ESP32-S3 microcontroller with integrated Wi-Fi for cloud connectivity. Audio input is captured using an INMP441 MEMS microphone and transmitted to cloud-based Al services such as Gemini or Open Al for speech-to-text conversion, intent detection, and contextual response generation. The backend also performs emotion analysis and text-to-speech synthesis. The processed response is delivered to the user through a 3W speaker driven by a MAX98357A amplifier, while a 1.3-inch TFT display renders a 3D humanoid avatar. A holographic acrylic pyramid creates a floating 3D visual effect.
The system is powered by a 2400mAh Li-ion battery with integrated charging and boost regulation. Experimental evaluation demonstrates stable operation, synchronized animation, and improved user engagement compared to traditional voice-only systems. The project provides a cost-effective and scalable solution for next-generation Al assistants.
Abstract
Hybrid Cloud Deployment Using AWS EC2 and On-Premises Virtual Machine
Nagaraj L, Dr. K. S. Gowrilakshmi
DOI: 10.17148/IJARCCE.2026.15382
Abstract: Cloud computing has fundamentally transformed how modern applications are built and deployed. This project presents a practical implementation of hybrid cloud architecture that integrates Amazon Web Services (AWS) Elastic Compute Cloud (EC2) with a locally hosted Oracle VirtualBox virtual machine running Ubuntu Linux. The system demonstrates how organizations can leverage the benefits of cloud infrastructure — including scalability, remote accessibility, and cost efficiency — while retaining on-premises control over sensitive workloads. The web application was deployed using the Apache HTTP Server on the AWS EC2 instance, with secure remote administration established via the SSH protocol. The hybrid architecture supports data backup and disaster recovery through cloud storage services (Amazon S3), distributed workload management via AWS Virtual Private Cloud (VPC), and system monitoring through AWS CloudWatch. Comprehensive unit, integration, system, and performance testing confirmed that all components function correctly. The project validates that hybrid cloud deployment is a practical, cost-effective approach for modern application hosting, and provides a reproducible reference implementation for cloud professionals and academic practitioners.
Keywords: Hybrid Cloud, AWS EC2, VirtualBox, Apache, SSH, Amazon S3, VPC, Ubuntu Linux, Virtualization, Cloud Deployment.
Keywords: Hybrid Cloud, AWS EC2, VirtualBox, Apache, SSH, Amazon S3, VPC, Ubuntu Linux, Virtualization, Cloud Deployment.
Abstract
CLOUD INTRUSION DETECTION SYSTEM
Mrs. K. Rajeswari, B. Thraya Gayathri, J. Hari Sravani, T. Himaja, N .P. Praharshita
DOI: 10.17148/IJARCCE.2026.15383
Abstract: The rapid adoption of cloud computing has significantly expanded the cybersecurity attack surface, exposing infrastructures to volumetric network attacks and zero-day exploits. Traditional signature-based Intrusion Detection Systems fail to detect novel threats and generate excessive false alarms, while heavy deep learning models introduce severe computational latency. This paper proposes a lightweight, real-time Cloud Intrusion Detection System, utilizing the unsupervised Isolation Forest machine learning algorithm. By continuously analyzing network telemetry such as packet transmission rates and failed logins, the model autonomously establishes a baseline of normal behavior and mathematically isolates statistical anomalies. The algorithmic backend is seamlessly integrated into a custom-built, interactive Security Operations Center (SOC) dashboard using the Streamlit framework, providing live 3D threat vector visualizations and automated PDF auditing. Experimental results confirm the system effectively intercepts simulated brute-force and DDoS payloads with minimal processing overhead, establishing a highly proactive cloud defense mechanism.
Keywords: Cloud Security, Intrusion Detection System, Isolation Forest, Anomaly Detection, Streamlit, SOC Dashboard, Unsupervised Learning.
Keywords: Cloud Security, Intrusion Detection System, Isolation Forest, Anomaly Detection, Streamlit, SOC Dashboard, Unsupervised Learning.
Abstract
Chatbot Development Using AWS Lex and Lambda
R. Sri Vignesh, Dr. B. Narasimhan
DOI: 10.17148/IJARCCE.2026.15384
Abstract: The proliferation of cloud-native services has created new possibilities for building intelligent conversational interfaces without the overhead of maintaining dedicated infrastructure. This paper presents the design, implementation, and evaluation of a customer-facing chatbot constructed using Amazon Lex V2 for natural language understanding and AWS Lambda for serverless fulfillment logic. The proposed system achieves an intent recognition accuracy of 94.2% across a test corpus of five hundred utterances, with an average end-to-end response latency of 310 milliseconds. The architecture integrates seamlessly with a suite of AWS backend services—DynamoDB, Simple Notification Service, and Simple Email Service—enabling context-aware, multi-turn conversations for a retail order-management use case. Comparative analysis against equivalent implementations on Google Dialogflow and IBM Watson demonstrates competitive accuracy and markedly lower operational cost at moderate traffic volumes. The paper discusses architectural trade-offs, cold-start mitigation strategies, slot validation patterns, and directions for future enhancement through sentiment analysis and personalization layers.
Keywords: Amazon Lex, AWS Lambda, serverless computing, natural language understanding, conversational AI, chatbot, cloud computing, intent recognition.
Keywords: Amazon Lex, AWS Lambda, serverless computing, natural language understanding, conversational AI, chatbot, cloud computing, intent recognition.
Abstract
Implement Auto Scaling with Docker and AWS
EC Auto Scaling, Nishok T N, Dr. S. Thavamani
DOI: 10.17148/IJARCCE.2026.15385
Abstract: Cloud computing has become an essential technology for modern software development and deployment. One of the key advantages of cloud computing is the ability to scale resources according to application demand. In traditional systems, servers are configured with fixed resources, which can cause performance problems when user traffic increases and resource wastage when traffic decreases. Auto scaling is a cloud computing feature that helps solve this issue by automatically increasing or decreasing computing resources based on workload.
This project focuses on implementing an auto scaling infrastructure using Docker containers and AWS EC2 instances. Docker is a containerization platform that allows developers to package applications and their dependencies into lightweight containers. These containers ensure that the application runs consistently across different environments. AWS EC2 provides virtual machines in the cloud that can host these containers.
In the proposed architecture, an Elastic Load Balancer distributes incoming traffic across multiple EC2 instances. Each instance runs a Docker container that hosts the application. AWS Auto Scaling monitors system metrics such as CPU utilization and automatically launches additional instances when the load increases. Similarly, when the load decreases, extra instances are terminated to optimize resource usage and reduce cost.
This system demonstrates how containerization and cloud auto scaling technologies can work together to create scalable, reliable, and cost‑efficient cloud applications.
This project focuses on implementing an auto scaling infrastructure using Docker containers and AWS EC2 instances. Docker is a containerization platform that allows developers to package applications and their dependencies into lightweight containers. These containers ensure that the application runs consistently across different environments. AWS EC2 provides virtual machines in the cloud that can host these containers.
In the proposed architecture, an Elastic Load Balancer distributes incoming traffic across multiple EC2 instances. Each instance runs a Docker container that hosts the application. AWS Auto Scaling monitors system metrics such as CPU utilization and automatically launches additional instances when the load increases. Similarly, when the load decreases, extra instances are terminated to optimize resource usage and reduce cost.
This system demonstrates how containerization and cloud auto scaling technologies can work together to create scalable, reliable, and cost‑efficient cloud applications.
Abstract
RFID And Fingerprint Based Digital Passport For Cross Border Authentication
Mr. H.M. Gaikwad, Mr. S.V. Waghmare, Saili Borse, Krutika Bhamare, Vidhi Daryani, Darshana Kothari
DOI: 10.17148/IJARCCE.2026.15386
Abstract: Advancements in technology have improved the verification of travel documents through the use of electronic passports (e-passports). These passports follow International Civil Aviation Organization standards and store biometric identifiers such as fingerprints. Biometric technology combined with RFID tags enhances identity verification and reduces the use of counterfeit documents. Fingerprint-based biometric e-passports help simplify travel processes while improving global security. They also focus on protecting the privacy and sensitive biometric data of passport holders. However, certain security challenges and vulnerabilities still exist in the system. This study analyzes the cryptographic features, biometric techniques, and security protocols used in e-passports. It also examines possible risks where individuals may attempt to bypass the system. In the proposed AI/ML-based crime monitoring system, registered crimes are analyzed and categorized by severity. If a serious crime is detected, the information is used during passport verification to evaluate the applicant’s criminal background before allowing travel[1].
Keywords: RFID Tag, RFID Reader, E-Passport.
Keywords: RFID Tag, RFID Reader, E-Passport.
Abstract
ROAD RULES SIMULATOR: AN INTERACTIVE LEARNING SYSTEM FOR TRAFFIC EDUCATION
Adil Niham, Goban Roshan P, Kiran Mohan, Melvin Alex, Prof. Marina Glastin
DOI: 10.17148/IJARCCE.2026.15387
Abstract: Road safety is still a major global concern, with many accidents being caused by inexperienced drivers,risky driving practices and lack of knowledge of traffic laws.Textbooks and lectures in the classroom are the mainstays of traditional driver education approaches, which frequently fall short of offering interesting and useful learning opportunities. In order to enhance traffic rule comprehension through interactive simulation and game-based learning, this study introduces the Road Rules Simulator, a gamified learner licensing training system. To provide an interesting training environment, the system combines scenario-driven driving simulations created with the Unity engine and C# programming with quiz-based learning. To improve student motivation and knowledge retention, the suggested system includes gamification components like levels, scoring, badges, awards, and real-time feedback. Additionally, it has analytics and progress tracking components that analyze student performance, identify infractions, and offer remedial advice. Without the dangers of the real world, the simulator offers a secure virtual setting where students can hone their driving abilities and follow traffic laws. Combining simulation and gamification enhances decision-making skills, engagement, and rule comprehension, according to experimental application. For contemporary driver education, the system offers an affordable, scalable, and accessible option.
Keywords: Road Rules Simulator, Driving Simulation, Gamified Learning, Traffic Rule Education, Driver Training System.
Keywords: Road Rules Simulator, Driving Simulation, Gamified Learning, Traffic Rule Education, Driver Training System.
Abstract
AI-Driven Second-Brain Knowledge Management System
Mrs. P. H. Nawale, Makarand Mangesh Ahire, Nikhil Nilesh Potdar, Aquino Harold Rodrigues, Shaikh Mohammed Zaid Faiyyaz
DOI: 10.17148/IJARCCE.2026.15388
Abstract: In the modern digital era, individuals interact with vast amounts of information daily, including academic notes, research materials, personal ideas, documents, and tasks. Managing this information efficiently is a major challenge due to scattered storage systems and the absence of intelligent organization tools. Traditional note-taking and productivity applications act only as passive storage platforms and do not provide intelligent reasoning or knowledge retrieval capabilities. This paper presents an AI-Driven Second-Brain Knowledge Management System, a cognitive digital platform designed to mimic the information storage and recall capabilities of the human brain. The system integrates Artificial Intelligence, Natural Language Processing (NLP), semantic embeddings, and knowledge graph technologies to automatically organize user information into an interconnected knowledge structure. The proposed system enables intelligent note linking, semantic search, AI-generated summaries, flashcard generation, and contextual recommendations. A vector database stores semantic embeddings of knowledge elements, allowing efficient retrieval based on meaning rather than simple keyword matching. The results demonstrate that the system significantly improves information retrieval speed, enhances knowledge connectivity, and reduces cognitive overload for users managing large volumes of information. The platform provides a scalable foundation for AI-assisted productivity, learning, and research workflows.
Keywords: Second Brain, Artificial Intelligence, Knowledge Management, Semantic Search, Vector Database, Knowledge Graph, Natural Language Processing.
Keywords: Second Brain, Artificial Intelligence, Knowledge Management, Semantic Search, Vector Database, Knowledge Graph, Natural Language Processing.
Abstract
AI based underwater Remotely Operated Vehicle (ROV) for Inspection and surveillance
Mr. H.M. Gaikwad, Vaibhav Bhusare, Rida Khan, Aarya Nagpure
DOI: 10.17148/IJARCCE.2026.15389
Abstract: Underwater inspection and surveillance are essential in marine research, dam structure monitoring, underwater pipeline inspection, defence surveillance, and search-and-rescue operations. Conventional inspection methods rely heavily on human divers, which involve high operational risk, limited working duration, and significant cost. Industrial- grade underwater robotic systems are available; however, they are often expensive and not easily accessible for academic or small-scale applications. This paper presents the design, development, and implementation of an Artificial Intelligence (AI) Based Underwater Remotely Operated Vehicle (ROV) intended for real-time surveillance and inspection. The system integrates a waterproof mechanical structure, multi-directional thruster system, embedded microcontroller-based control unit, high-definition underwater camera, and an AI-based object detection module. The proposed ROV enables stable underwater navigation, real-time video transmission, and intelligent detection of underwater objects and anomalies. The hardware prototype is developed at an approximate cost of â‚ą60,000, making it significantly more affordable than industrial systems. Experimental testing in shallow freshwater environments demonstrates reliable manoeuvrability, stable video streaming, and effective AI-assisted monitoring.
Keywords: Underwater ROV, Artificial Intelligence, Object Detection, Embedded Systems, Surveillance, Marine Robotics, Real-Time Monitoring.
Keywords: Underwater ROV, Artificial Intelligence, Object Detection, Embedded Systems, Surveillance, Marine Robotics, Real-Time Monitoring.
Abstract
Adaptive Learning Technologies in Modern Education
Madhuri Ramdas Gagare, Sakshi Dhananjay Jadhav
DOI: 10.17148/IJARCCE.2026.15390
Abstract: Adaptive learning technologies are transforming modern education by providing personalized learning experiences for students. Traditional teaching methods usually follow a fixed curriculum and the same pace for all learners, which may not suit every student’s learning ability and style. Adaptive learning systems use technologies such as Artificial Intelligence (AI), machine learning, and learning analytics to analyze student performance, behavior, and progress. Based on this data, the system adjusts the difficulty level, learning materials, and teaching methods according to the individual needs of each student. These technologies help students learn at their own pace and receive instant feedback, which improves understanding and engagement. Adaptive learning platforms are widely used in online education systems, digital classrooms, and e-learning applications. They also assist teachers by tracking student progress and identifying areas where students need additional support.However, the implementation of adaptive learning technologies also presents certain challenges, including high development costs, data privacy concerns, and dependence on digital infrastructure. Despite these challenges, adaptive learning technologies have significant potential to improve the quality, accessibility, and effectiveness of modern education. As technology continues to advance, adaptive learning systems are expected to play a key role in shaping the future of education by creating more personalized and efficient learning environments.
Keywords: Adaptive Learning, Artificial Intelligence, Personalized Learning, E-Learning Platforms, Learning Analytics, Intelligent Tutoring Systems, Educational Technology.
Keywords: Adaptive Learning, Artificial Intelligence, Personalized Learning, E-Learning Platforms, Learning Analytics, Intelligent Tutoring Systems, Educational Technology.
Abstract
An Automated Human Violence Detection for Surveillance using Deep Learning
Mr. H.M. Gaikwad, Sharvari Gangurde, Priyanka Moule, Sejal Patil, Aarya Patil
DOI: 10.17148/IJARCCE.2026.15391
Abstract: In recent years, ensuring public safety in crowded environments such as transportation hubs, campuses, and urban areas has become increasingly challenging due to the rise in violent incidents. Traditional surveillance systems rely on continuous human observation, which is inefficient, time-consuming, and prone to errors. To address these issues, this project proposes an AI-based automated surveillance system capable of detecting violent human activities in real time using deep learning techniques. The proposed system utilizes the YOLOv8 Nano model developed by Ultralytics along with a custom-trained dataset to identify violent behavior directly from video streams captured through live webcams (CCTV cameras). The YOLO-based approach performs fast, single stage object detection, enabling real-time monitoring with lower computational requirements. By learning spatial patterns associated with aggressive actions, the system can effectively differentiate between violent and non-violent activities even in dynamic and crowded environments. The developed framework is integrated with a monitoring interface that generates instant alerts whenever violent behavior is detected, thereby assisting authorities in taking timely action. Experimental results indicate that the system achieves reliable detection speed and satisfactory accuracy, making it suitable for deployment in smart surveillance systems, institutional security setups, and public safety monitoring applications [1].
Keywords: Artificial Intelligence, Machine Learning, Deep Learning, Computer Vision, Object Detection, Human Activity Recognition, Violence Detection, Smart Surveillance, Real-time Monitoring, Security Automation, Public Safety System, Emergency Alert System, Video Analytics, Image Processing, Feature Extraction, Convolutional Neural Networks, YOLO Algorithm, Dataset Training, Model Evaluation, Accuracy Optimization, Intelligent Monitoring, AI- based Security, Threat Detection, Smart City Safety, Automated Surveillance.
Keywords: Artificial Intelligence, Machine Learning, Deep Learning, Computer Vision, Object Detection, Human Activity Recognition, Violence Detection, Smart Surveillance, Real-time Monitoring, Security Automation, Public Safety System, Emergency Alert System, Video Analytics, Image Processing, Feature Extraction, Convolutional Neural Networks, YOLO Algorithm, Dataset Training, Model Evaluation, Accuracy Optimization, Intelligent Monitoring, AI- based Security, Threat Detection, Smart City Safety, Automated Surveillance.
Abstract
AgroBot: AI-Powered Smart Robotic Lawn Mower
Mr. H.M. Gaikwad, Rutuja Yogesh Deshmukh, Shubham Dnyaneshwar Kunde
DOI: 10.17148/IJARCCE.2026.15392
Abstract: Autonomous lawn mowers represent a significant advancement in agricultural robotics, offering labor reduction and precision maintenance for residential and commercial applications. This paper presents the design, implementation, and field validation of AgroBot, a low-cost autonomous lawn mower incorporating GPS navigation, LiDAR-based obstacle avoidance, and AI-powered weed detection capabilities. The system employs a Raspberry Pi 4 as the primary computing platform running ROS Noetic, integrated with Pixhawk 2.4.8 for motion control and sensor fusion. Field testing conducted over a 500 m² area demonstrated 90.0% average coverage efficiency across five missions, with robust obstacle avoidance (100% success rate) and AI-based weed detection achieving 87.3% mean Average Precision (mAP). The complete system cost of ₹58,450 represents a 51-71% cost reduction compared to commercial alternatives while providing superior customizability and research extensibility. Performance metrics include 150-minute operational runtime, 0.85m GPS-EKF positioning accuracy, and autonomous operation through Boustrophedon path planning with real-time dynamic obstacle avoidance.
Keywords: Autonomous Navigation, Lawn Mower Robot, SLAM, YOLOv5, Coverage Path Planning, ROS, Sensor Fusion, Mobile Robotics, Agricultural Robotics
Keywords: Autonomous Navigation, Lawn Mower Robot, SLAM, YOLOv5, Coverage Path Planning, ROS, Sensor Fusion, Mobile Robotics, Agricultural Robotics
Abstract
Megatron: AI Powered Navigation And Information Bot
Mr. H.M. Gaikwad, Ms. V. N. Lawand, Krishi Jadhav, Abhirav Pawar, Vishal Dhule
DOI: 10.17148/IJARCCE.2026.15393
Abstract: The rapid growth of Artificial Intelligence and machine learning technologies has significantly improved the ability of machines to understand and respond to human commands. Intelligent navigation and information systems have become increasingly important in environments such as educational institutions, public spaces, and smart campuses. In this project, the system will assist users by providing navigation guidance and information through voice-based interaction. The proposed system, called Megatron: AI Powered Navigation and Information Bot, uses Artificial Intelligence techniques to understand voice commands and provide accurate responses and directions to users.
The system integrates speech recognition, natural language processing, and intelligent navigation to create an interactive and user-friendly assistant. Voice commands are captured using a microphone and processed using speech recognition algorithms to identify user queries. The system then interprets the request and provides appropriate navigation instructions or relevant information about locations and services.
This project demonstrates an important application of Artificial Intelligence technology, which enables efficient human– machine interaction and improves accessibility for users. The proposed system aims to simplify navigation and information retrieval in complex environments by providing real-time assistance. By combining voice recognition, machine learning, and intelligent navigation techniques, the system enhances user convenience and supports the development of smart and automated environments.
Keywords: Artificial Intelligence, Voice Recognition, Navigation System, Information Bot, Natural Language Processing, Smart Assistance
The system integrates speech recognition, natural language processing, and intelligent navigation to create an interactive and user-friendly assistant. Voice commands are captured using a microphone and processed using speech recognition algorithms to identify user queries. The system then interprets the request and provides appropriate navigation instructions or relevant information about locations and services.
This project demonstrates an important application of Artificial Intelligence technology, which enables efficient human– machine interaction and improves accessibility for users. The proposed system aims to simplify navigation and information retrieval in complex environments by providing real-time assistance. By combining voice recognition, machine learning, and intelligent navigation techniques, the system enhances user convenience and supports the development of smart and automated environments.
Keywords: Artificial Intelligence, Voice Recognition, Navigation System, Information Bot, Natural Language Processing, Smart Assistance
Abstract
Development of Robust Trust Management Framework for Medical IoT
K Chakravarthy Bheri, Manas Kumar Y
DOI: 10.17148/IJARCCE.2026.15394
Abstract: Modern healthcare without IoT devices is un imaginable. They've diffused into almost every corner of patient monitoring—tracking heart rates, oxygen levels, you name it. And honestly, things run a lot smoother because of them. But this shift hasn't come without problems—security and trust have become major concerns. To tackle this, we built a system that brings together blockchain and deep learning to keep an eye on devices in real time and figure out which ones can be trusted. The system uses three different deep learning models working together to constantly evaluate device trust. One looks for anomalies using LSTM, another analyzes behavior with CNN, and a third predicts threats through GRU. All of this is backed by a simple but effective blockchain that keeps a tamper-proof record of trust scores and transactions—making things transparent and easy to audit Healthcare administrators can monitor device status, see trust distributions, handle alerts, and investigate blockchain transactions using the system's interactive graphical user interface, which was created with Tkinter. Effective trust score computation across various device types, including the heart, is demonstrated by experimental results
Keywords: blockchain, deep learning, healthcare security, Internet of Things, trust management
Keywords: blockchain, deep learning, healthcare security, Internet of Things, trust management
Abstract
DriveMate AI: Personalized Intelligent Driver Assistant
Emmanuel SJ, Devnandhan M, Joel Roy, Shan Jahan, Assistant Prof. Ancy Das
DOI: 10.17148/IJARCCE.2026.15395
Abstract: Driver fatigue and distraction are among the leading causes of road accidents worldwide. With the increasing number of vehicles on the road, ensuring driver safety has become a major concern. Many existing driver assistance systems focus mainly on vehicle control rather than monitoring the driver's physical and mental condition.
This project presents DriveMate AI: Personalized Intelligent Driver Assistant, an intelligent system designed to improve road safety using artificial intelligence and computer vision technologies. The system continuously monitors the driver using a camera and analyzes facial features such as eye movements, blink rate, and head position to detect signs of fatigue and drowsiness. When unsafe conditions are detected, the system generates real-time alerts to warn the driver.
In addition to fatigue detection, the system also includes emotion recognition and conversational AI modules to interact with the driver and reduce boredom during long journeys. The system can also detect potential collision risks using object detection techniques and trigger warning alerts.
The proposed DriveMate AI system integrates multiple modules including driver monitoring, emotion detection, chatbot interaction, collision prediction, and emergency response. The system aims to create a personalized driver assistance experience that improves driving safety and comfort.
Keywords: Driver Monitoring System, Artificial Intelligence, Computer Vision, Drowsiness Detection, Emotion Recognition, Road Safety.
This project presents DriveMate AI: Personalized Intelligent Driver Assistant, an intelligent system designed to improve road safety using artificial intelligence and computer vision technologies. The system continuously monitors the driver using a camera and analyzes facial features such as eye movements, blink rate, and head position to detect signs of fatigue and drowsiness. When unsafe conditions are detected, the system generates real-time alerts to warn the driver.
In addition to fatigue detection, the system also includes emotion recognition and conversational AI modules to interact with the driver and reduce boredom during long journeys. The system can also detect potential collision risks using object detection techniques and trigger warning alerts.
The proposed DriveMate AI system integrates multiple modules including driver monitoring, emotion detection, chatbot interaction, collision prediction, and emergency response. The system aims to create a personalized driver assistance experience that improves driving safety and comfort.
Keywords: Driver Monitoring System, Artificial Intelligence, Computer Vision, Drowsiness Detection, Emotion Recognition, Road Safety.
Abstract
Network Intrusion Detection System Using Machine Learning
Anusree P S, Aryanandha Anil R, Fathima S, Gopika K and Prof. Merlin K Thomas
DOI: 10.17148/IJARCCE.2026.15396
Abstract: With the rapid growth of Internet services and digital communication technologies, cyberattacks have become increasingly frequent, complex, and sophisticated. Modern computer networks are continuously exposed to various types of security threats that can compromise the confidentiality, integrity, and availability of data. Traditional security mechanisms such as firewalls and signature-based detection systems are widely used to protect networks; however, these methods are often ineffective in detecting unknown or evolving attack patterns. Therefore, more intelligent and adaptive security solutions are required to ensure effective protection of network infrastructures.
Intrusion Detection Systems (IDS) play an important role in identifying malicious activities by continuously monitoring network traffic and detecting abnormal behaviour. In recent years, machine learning techniques have gained significant attention in the field of cybersecurity due to their ability to analyse large volumes of data and identify hidden patterns associated with cyberattacks. This paper proposes a Machine Learning-based Network Intrusion Detection System designed to improve the detection of cyberattacks in network environments. The proposed system performs several processes including data preprocessing, feature selection, and classification to distinguish between normal and malicious network traffic. Various machine learning algorithms are applied to analyse network behaviour and detect attacks such as Denial of Service (DoS), Probe attacks, and unauthorized access attempts.
The proposed model is trained and evaluated using a standard intrusion detection dataset and implemented using Python. Experimental analysis demonstrates that the proposed approach improves detection accuracy while reducing false alarm rates. The results indicate that the system provides an efficient and reliable solution for enhancing network security and detecting cyber threats in modern computing environments.
Keywords: Intrusion Detection System, Machine Learning, Cyber Security, Network Security, Data Classification
Intrusion Detection Systems (IDS) play an important role in identifying malicious activities by continuously monitoring network traffic and detecting abnormal behaviour. In recent years, machine learning techniques have gained significant attention in the field of cybersecurity due to their ability to analyse large volumes of data and identify hidden patterns associated with cyberattacks. This paper proposes a Machine Learning-based Network Intrusion Detection System designed to improve the detection of cyberattacks in network environments. The proposed system performs several processes including data preprocessing, feature selection, and classification to distinguish between normal and malicious network traffic. Various machine learning algorithms are applied to analyse network behaviour and detect attacks such as Denial of Service (DoS), Probe attacks, and unauthorized access attempts.
The proposed model is trained and evaluated using a standard intrusion detection dataset and implemented using Python. Experimental analysis demonstrates that the proposed approach improves detection accuracy while reducing false alarm rates. The results indicate that the system provides an efficient and reliable solution for enhancing network security and detecting cyber threats in modern computing environments.
Keywords: Intrusion Detection System, Machine Learning, Cyber Security, Network Security, Data Classification
Abstract
PERF CLOUD: A Hybrid DTW–GRU Framework for Black-Box Virtual Machine Performance Degradation Prediction in Multi-Tenant Public Clouds
Aiswarya Lekshmi JS, Aiswarya Vijayan, Ardra Vinod, Mathews Franco, Merlin K Thomas, Ajeesh S
DOI: 10.17148/IJARCCE.2026.15397
Abstract: Cloud computing environments host multiple virtual machines (VMs) on shared physical infrastructure, leading to unpredictable performance due to resource contention and workload variability. In public cloud settings, virtual machines operate as black-box systems, restricting access to internal application metrics and making accurate performance prediction highly challenging. To address this issue, this paper proposes PERFCLOUD, a hybrid machine learning framework for performance degradation prediction in multi-tenant public clouds. The proposed approach consists of three stages: (i) application type identification using Dynamic Time Warping (DTW) to align time-series workload patterns, (ii) highly correlated metric selection using Pearson Correlation to reduce noise and improve feature relevance, and (iii) time-series performance forecasting using a Gated Recurrent Unit (GRU) neural network. The model predicts both application type and VM performance status as Best or Degraded, enabling proactive resource management and interference mitigation. Experimental evaluation on a real-world cloud workload dataset demonstrates that the proposed DTW–GRU hybrid model significantly outperforms baseline LSTM-based approaches, achieving an accuracy of 99% with reduced computational overhead. The framework is further integrated into a web-based monitoring system for real-time VM performance analysis. The results validate the effectiveness, scalability, and adaptability of the proposed solution for intelligent cloud resource optimization.
Keywords: Cloud Performance Prediction, Black-Box Virtual Machines, Dynamic Time Warping (DTW), Gated Recurrent Unit (GRU), Performance Degradation Detection, Multi-Tenant Public Cloud, Machine Learning-Based Resource Optimization
Keywords: Cloud Performance Prediction, Black-Box Virtual Machines, Dynamic Time Warping (DTW), Gated Recurrent Unit (GRU), Performance Degradation Detection, Multi-Tenant Public Cloud, Machine Learning-Based Resource Optimization
Abstract
FarmSphere: A Smart Digital Platform for Sustainable and Connected Agriculture
Avishka Gaykar, Gauri Shelke, Bhavesh Suryavanshi, Mrs. Kalyani Kapde
DOI: 10.17148/IJARCCE.2026.15398
Abstract: FarmSphere is a web-based platform designed to bring farmers, buyers, and agriculture learners together through a single digital space. FarmSphere is a platform that enables farmers to list and sell a variety of products, including fruits, vegetables, grains, herbs, and flowers. Buyers can easily browse, search, and purchase these products directly from the platform. In addition to the product marketplace, FarmSphere features a Learning Hub, which offers articles on sustainable farming practices, irrigation techniques, pest control strategies, and government schemes. A Waste Utilisation module helps farmers understand how agricultural waste can be reused productively. The system uses Firebase as its backend, which handles user authentication, data storage, image hosting, and deployment. This paper discusses the design, architecture, features, and technologies used in FarmSphere, as well as its potential to make agriculture more accessible and sustainable for small and medium-sized farmers.
Keywords: Smart Agriculture, Digital Farming Platform, Agricultural Marketplace, Firebase, Sustainable Farming, Web Application, E-Commerce, Farmer Empowerment
Keywords: Smart Agriculture, Digital Farming Platform, Agricultural Marketplace, Firebase, Sustainable Farming, Web Application, E-Commerce, Farmer Empowerment
Abstract
Football Match Analysis With Yolo
Sachin Yadav, Vipul Chavan, Janmesh Vyas, Dr. M. M. Deshpande
DOI: 10.17148/IJARCCE.2026.15399
Abstract: Football match analysis is an important task in modern sports analytics for evaluating player performance and tactical strategies. Traditional analysis methods depend on manual observation, which is time-consuming and subjective. This paper presents an automated football match analysis system using the YOLO (You Only Look Once) deep learning algorithm for real-time object detection. The proposed system processes football match videos to detect key entities such as players, referees, and the ball. YOLO performs single-stage detection by predicting bounding boxes and class probabilities in a single forward pass, enabling fast and efficient analysis.The model is trained on annotated football match datasets and evaluated using standard metrics including precision, recall, mean Average Precision (mAP), and frames per second (FPS). Experimental results show that the system achieves high detection accuracy with real-time performance, even in dynamic match conditions. The proposed approach reduces computational complexity compared to traditional methods and can assist coaches and analysts in performance evaluation and tactical decision- making.
Abstract
MediSphere: AI-Driven Diagnostics with Precision Scoring and Therapeutic Intelligence
Dr. Jagdish.W. Bakal, Rohan Hanmant Saste, Kishor Ramesh Shirke, Abhay Rajendra Jadhav, Gaurav Ananda Lende
DOI: 10.17148/IJARCCE.2026.153100
Abstract: Medisphere is an AI-powered web-based platform designed to enhance diagnostic accuracy and support clinical decision-making through precision scoring. The platform utilizes advanced machine learning models, trained on a wide range of medical datasets, to interpret patient symptoms, clinical history, and diagnostic test outcomes. By producing a diagnostic precision score, Medisphere supports healthcare professionals in assessing the probability of particular conditions, minimizing diagnostic errors, and enhancing patient care. The website integrates an intuitive user interface for physicians and medical staff, enabling seamless data input, visualization of diagnostic insights, and comparative analysis with standard clinical guidelines. Medisphere’s AI-driven scoring framework aims to act as a supportive tool rather than a replacement for medical expertise, providing transparent, ex- plainable insights that augment physician judgment. Ultimately, the project demonstrates the transformative role of AI in healthcare by combining scalability, precision, and accessibility to advance early detection, personalized treatment, and overall diagnostic efficiency.
Keywords: Medisphere, AI-powered healthcare, diagnostic ac- curacy, precision scoring, ma chine learning models, clinical decision support, personalized treatment.
Keywords: Medisphere, AI-powered healthcare, diagnostic ac- curacy, precision scoring, ma chine learning models, clinical decision support, personalized treatment.
Abstract
Timex: Timetable Automation System for Educational Resource Scheduling in Smart Campus
Prof. Vanchala Sutar, Ishita Tarak Mehta, Darshan Deepak Redekar, Aniket Ramchandra Shinde
DOI: 10.17148/IJARCCE.2026.153101
Abstract: Academic timetable generation is a complex scheduling problem involving multiple institutional resources such as faculty members, classrooms, laboratories, student batches, and time slots. Traditional manual scheduling approaches are time-consuming and often lead to conflicts, inefficient resource utilization, and administrative overhead. This paper proposes Timex, an intelligent timetable automation framework designed for smart campus environments. The system integrates constraint-based scheduling with artificial intelligence–inspired optimization techniques to efficiently allocate academic resources while satisfying institutional policies and operational constraints. The proposed system supports dynamic timetable generation, real-time schedule modifications, and automated conflict detection.
Experimental evaluation demonstrates that the Timex framework significantly improves classroom and laboratory utilization while reducing scheduling time compared with traditional manual approaches. The system provides a scalable and reliable solution for modern educational institutions seeking intelligent academic resource management.
Keywords: Academic Timetable Scheduling, Smart Campus, Artificial Intelligence, Constraint Satisfaction, Resource Optimization, Educational Automation
Experimental evaluation demonstrates that the Timex framework significantly improves classroom and laboratory utilization while reducing scheduling time compared with traditional manual approaches. The system provides a scalable and reliable solution for modern educational institutions seeking intelligent academic resource management.
Keywords: Academic Timetable Scheduling, Smart Campus, Artificial Intelligence, Constraint Satisfaction, Resource Optimization, Educational Automation
Abstract
MediScan AI: An Intelligent Vision-Based Platform for Prescription Decoding and Comprehensive Medicine Information Delivery
Mr. H.M. Gaikwad, Ms. V. N. Lawand, Vedant Dhotre, Krushna Nikam
DOI: 10.17148/IJARCCE.2026.153102
Abstract: In the modern healthcare ecosystem, deciphering handwritten prescriptions remains a critical patient safety challenge, particularly in developing nations and small clinical settings. Studies indicate that up to 61% of prescription errors are attributed to illegible handwriting, leading to severe health and economic consequences.To address this issue, this paper presents MediScan AI, a cloud-deployed web application that utilizes advanced multimodal artificial intelligence to analyze prescription images and deliver patient-friendly medicine information. The system employs a hybrid AI-database architecture, integrating Google Gemini Vision AI for accurate text extraction with a curated local database for validated medical data.A key innovation of the platform is the mandatory extraction and presentation of side effects and overdose risks for all identified medications. Experimental results demonstrate high accuracy across both printed and handwritten prescriptions, with an average processing time under 15 seconds. MediScan AI effectively bridges the gap between complex medical prescriptions and patient understanding, enhancing medication safety and accessibility.
Keywords: Artificial Intelligence, Prescription Decoding, OCR, Google Gemini Vision, Medication Safety, Hybrid Architecture
Keywords: Artificial Intelligence, Prescription Decoding, OCR, Google Gemini Vision, Medication Safety, Hybrid Architecture
Abstract
Real-Time Expense Tracker with Smart Budget Recommendation System
Nita Ingale, Nutan Avhad, Diksha Jadhav, Pratik Pawar, Soham Gawde
DOI: 10.17148/IJARCCE.2026.153103
Abstract: This paper presents a smart expense tracking and budget suggestion system designed to help users monitor spending habits and improve financial planning. The system provides an automated approach to recording daily expenses, categorizing transactions, and analyzing spending patterns in real time. User expense data is processed through a structured pipeline that includes data collection, preprocessing, classification, and budget analysis.
Expenses are categorized into predefined domains such as food, travel, utilities, and entertainment using rule-based and machine learning techniques. Based on historical spending behavior and monthly income, the system generates personalized budget recommendations and alerts users when spending exceeds predefined limits. Visual analytics in the form of charts and summaries enable users to gain clear insights into their financial behavior. The application emphasizes usability, accuracy, and real- time feedback while ensuring data security and privacy. The proposed system demonstrates an effective solution for personal financial management by promoting disciplined spending and informed budgeting decisions.
Keywords: Expense Tracker, Budget Management, Personal Finance, Spending Analysis, Financial Analytics, Budget Recommendation System
Expenses are categorized into predefined domains such as food, travel, utilities, and entertainment using rule-based and machine learning techniques. Based on historical spending behavior and monthly income, the system generates personalized budget recommendations and alerts users when spending exceeds predefined limits. Visual analytics in the form of charts and summaries enable users to gain clear insights into their financial behavior. The application emphasizes usability, accuracy, and real- time feedback while ensuring data security and privacy. The proposed system demonstrates an effective solution for personal financial management by promoting disciplined spending and informed budgeting decisions.
Keywords: Expense Tracker, Budget Management, Personal Finance, Spending Analysis, Financial Analytics, Budget Recommendation System
Abstract
Hybrid AI-Powered Legal Assistant for BNS Laws Using Structured Retrieval and Contextual Language Model Generation
Omkar Rajaram Barad, Atharava Atul Desai, Atharav Sachchidanand Bhogate, Saurabh Namdev Dhokare, Prof. (Dr.) Suresh Shirgave
DOI: 10.17148/IJARCCE.2026.153104
Abstract: Access to accurate legal information remains a major challenge for non-experts due to the complexity of statutory language and the evolving structure of Indian criminal law. This study proposes a hybrid AI-powered legal assistant that combines deterministic statutory retrieval with contextual large language model generation to provide reliable and interpretable legal explanations. The system integrates a structured legal database containing codified criminal statutes with a contextual explanation engine powered by a locally deployed large language model (LLaMA 3). A dual-layer architecture was implemented: an offline rule-based retrieval module for precise statutory matching and an online generative module for extended legal explanation. Query preprocessing includes keyword expansion, stemming, and multilingual normalization. Performance was evaluated using a curated dataset of 250 legal queries covering direct section lookups and semantic intent-based questions. The offline module achieved high precision for direct statutory queries with low latency suitable for interactive use. The contextual module improved interpretability and semantic coverage for intent-based questions. The hybrid strategy reduced hallucination risk compared to standalone generation by grounding responses in verified statutory content. The proposed hybrid legal assistant provides a scalable and accurate framework for legal information access. By combining deterministic retrieval and contextual generation, the system mitigates hallucination risks associated with large language models while enhancing user comprehension. The framework is designed to be extensible to multilingual and cross-jurisdictional applications.
Index Terms—Artificial Intelligence, Legal Informatics, Hybrid Retrieval, Large Language Models, Indian Criminal Law, Multilingual NLP
Index Terms—Artificial Intelligence, Legal Informatics, Hybrid Retrieval, Large Language Models, Indian Criminal Law, Multilingual NLP
Abstract
AI-Driven Ferry Booking and Assistance System for Smart and Inclusive Railway Services
Viren S. Adgale, Kashish P. Jadhav, Srushti P. Jadhav, Samiksha S. Herwade, Rushikesh B. Gherade
DOI: 10.17148/IJARCCE.2026.153105
Abstract: This paper presents an AI-based system designed to automate and optimize luggage ferry and passenger assistance services within railway stations, a domain that currently suffers from manual coordination, service delays, and accessibility challenges. These inefficiencies often lead to poor passenger experience, operational bottlenecks, and increased workload on station staff, particularly affecting elderly and differently-abled passengers. The proposed system monitors service requests and operational data during an initial execution phase and builds an intelligent service management profile using artificial intelligence techniques. Based on this learned behavior, the system predicts optimal driver assignments and service routes while providing real-time tracking and timely notifications to improve service delivery. The objective of this work is to enhance passenger satisfaction, improve operational efficiency, and promote inclusive railway services while maintaining system reliability and minimizing manual intervention.
Keywords: AI-based service management, railway assistance system, luggage ferry booking, wheelchair assistance, intelligent recommendation system, chatbot, smart transportation.
Keywords: AI-based service management, railway assistance system, luggage ferry booking, wheelchair assistance, intelligent recommendation system, chatbot, smart transportation.
Abstract
IFOS: AN ML-POWERED REAL-TIME INTELLIGENT FILE ORGANIZATION SYSTEM
Priyanka Balaso Ingale, Mayur Narendra Gavali, Third Year B.Tech, Computer Science Engineering (Artificial Intelligence), DKTE Ichalkaranji., Final Year B.Tech, Computer Science Engineering (Artificial Intelligence), DKTE Ichalkaranji
DOI: 10.17148/IJARCCE.2026.153106
Abstract: This paper presents the Intelligent File Organization System (IFOS), a lightweight supervised machine learning solution that automatically classifies and sorts files into predefined folders the moment they are downloaded. On average, a typical user downloads over 500 files per month across diverse categories PDFs, images, videos, archives, executables all accumulating in a single downloads folder with no automatic organization. When IFOS detects a new file, it extracts a 13-dimensional feature vector comprising file extension, MIME type, binary magic number bytes (first 8 bytes), log-transformed file size, and Shannon entropy, and passes it to a trained Random Forest classifier. The predicted category triggers an automated file-move operation into the corresponding pre-existing directory, requiring zero user intervention. Unlike rule-based tools that fail on mislabeled or extension-less files, IFOS is content-aware and achieves 97.4% classification accuracy and 87 ms average end-to-end latency across a balanced dataset of 10,000 real-world files spanning 10 categories.
Keywords: Machine Learning, File Classification, Intelligent File Organization, MIME Type Detection, Magic Numbers, Shannon Entropy, Random Forest, File System Automation, Real-Time Monitoring, Download Manager, Supervised Learning, Edge Case Robustness, Cross-Platform Automation
Keywords: Machine Learning, File Classification, Intelligent File Organization, MIME Type Detection, Magic Numbers, Shannon Entropy, Random Forest, File System Automation, Real-Time Monitoring, Download Manager, Supervised Learning, Edge Case Robustness, Cross-Platform Automation
Abstract
Self-Evolving Cognitive Architectures: Bridging Neural Optimization and Consciousness Simulation for AGI
Adhithyan A, Vivek S, Febin K James, Muhammed Abnan, Asst. Prof. Jinu L
DOI: 10.17148/IJARCCE.2026.153107
Abstract: This paper presents the design and theoretical foundations of a Self-Evolving Cognitive Architecture (SECA), a unified framework that combines autonomous neural architecture evolution with consciousness-inspired cognitive mod- eling. The proposed system leverages evolutionary algorithms, particularly customized genetic operators, to iteratively redesign and optimize neural network architectures without human intervention. In parallel, it incorporates Theory of Mind (ToM) principles, recursive self-modeling, and agent-based logic to simulate subjective awareness, emergent be- havior, and adaptive reasoning. Conventional approaches to artificial intelligence typically emphasize either structural optimization, as seen in automated architecture search, or cognitive emulation, as pursued in consciousness simulators. SECA addresses this divide by providing a dual layered paradigm in which neural substrates evolve continuously while cognitive processes develop recursively. This synthesis offers both computational adaptability and higher order reasoning capabilities, enabling AI systems to dynamically restructure their architectures while simultaneously reflecting on their internal states, predicting the intentions of other agents, and adapting behavior to complex, dynamic environments. By integrating architectural self-optimization with consciousness simulation, SECA represents a significant advancement to- ward Artificial General Intelligence (AGI). The framework facilitates the creation of AI systems that are not only efficient and scalable but also capable of introspection, adaptive learning, and human-like social cognition.
Keywords: Self-Evolving Neural Architectures, Consciousness Simulation, Theory of Mind, Evolutionary Algorithms, AGI, Adaptive AI
Keywords: Self-Evolving Neural Architectures, Consciousness Simulation, Theory of Mind, Evolutionary Algorithms, AGI, Adaptive AI
Abstract
Advanced Marketplace Suggestion System For Farm Produces Using AIML
Atharv C. Chavan, Manav P. Bansode, Om L. Kamble, Rahul N. Kamble, Vaijanath M. Wale, Prof. D. D. Mahajan
DOI: 10.17148/IJARCCE.2026.153108
Abstract: This paper presents an Advanced Marketplace Suggestion System for Farm Producers designed to enhance decision-making in agricultural trade through Artificial Intelligence (AI) integration. Traditional agricultural market systems often fail to differentiate produce based on quality, resulting in inefficient distribution and reduced profitability for farmers. The proposed system addresses this challenge by incorporating AI-based crop quality classification to categorize agricultural produce into four levels: Premium, Best, Good, and Bad. Based on the identified quality, the system intelligently recommends appropriate marketplaces such as exporters, processing industries, or local markets.
Keywords: Artificial Intelligence, Crop Quality Classification, Marketplace Recommendation System, Machine Learning, Smart Farming.
Keywords: Artificial Intelligence, Crop Quality Classification, Marketplace Recommendation System, Machine Learning, Smart Farming.
Abstract
Roamio: Leveraging Artificial Intelligence for Smart Traveler Matching, Itinerary Generation, and Solo Travel Safety
Vidhi Sankhe, Richa Bhardwaj, Tanishq Tamhankar, Abhay Sonkawade, Mrs. Ashwini More
DOI: 10.17148/IJARCCE.2026.153109
Abstract: The rise of social networking tools and mobile technologies has greatly influenced modern-day travel. Roamio is a full-stack cross-platform application that facilitates connections between compatible travel partners for solo travellers using AI-powered preference matching, real-time chat, intelligent trip planning, and SOS safety features. This paper outlines the design, development, implementation, and evaluation of Roamio using Visual Studio Code as the primary code editor, Expo React Native as the cross-platform mobile application development tool, and MongoDB as the NoSQL database backend. The application server uses Node.js/Express, MongoDB Atlas as a persistent cloud database, and various AI tools using LLM APIs. A cosine similarity algorithm is used to match users based on a multi-dimensional preference vector. A curated AI travel feed provides personalized travel recommendations to users, while an SOS feature allows users to send distress messages to up to five saved contacts. The results show that using AI to personalize travel recommendations with community-driven social networking for travel greatly reduces social friction for solo travellers.
Keywords: Travel companion platform, Expo React Native, MongoDB, AI trip planning, emergency SOS, social travel, preference matching, cross-platform mobile application
Keywords: Travel companion platform, Expo React Native, MongoDB, AI trip planning, emergency SOS, social travel, preference matching, cross-platform mobile application
Abstract
Comparison Of Anthropometric Variable With Respect To Forearm Muscle Circumference Among Tribal And Non-Tribal Sportsmen With Reference To Age Groups
Gaude Pralay Rohidas
DOI: 10.17148/IJARCCE.2026.153110
Abstract: Anthropometric characteristics play an important role in determining physical fitness and sports performance. Among these characteristics, forearm muscle circumference is considered an important indicator of upper limb strength and muscular endurance. The purpose of the present study was to compare forearm muscle circumference between tribal and non-tribal sportsmen of Goa with reference to two age groups (21–25 years and 26–30 years). A total of 100 male sportsmen were selected as subjects for the study, including 50 tribal and 50 non-tribal players from various sports disciplines. The subjects were further categorized into two age groups. Forearm muscle circumference was measured using a standard measuring tape following accepted anthropometric procedures. The collected data were analyzed using mean, standard deviation, and independent t-test to determine the significance of difference between groups. The results revealed that non-tribal sportsmen demonstrated slightly higher mean values of forearm muscle circumference compared to tribal sportsmen in both age groups; however, the differences were not statistically significant at the 0.05 level of significance. The findings suggest that regular sports participation contributes to similar muscular development of the forearm among tribal and non-tribal athletes regardless of age category.
Keywords: ; Anthropometry, Forearm circumference, Tribal athletes, Non-tribal athletes, Age groups, Sports science
Keywords: ; Anthropometry, Forearm circumference, Tribal athletes, Non-tribal athletes, Age groups, Sports science
Abstract
Adaptive Traffic Signal Control System Based on Future Traffic Environment
Parshwa Pravin Gebise, Aditya Shailesh Kulkarni, Devendra Santosh Gurav, Siddharth Deepak Jawale, Shivam Sanjay Dafale
DOI: 10.17148/IJARCCE.2026.153111
Abstract: Traffic congestion is a major challenge in urban environments due to increasing vehicle density and inefficient fixed-time traffic signal systems. Traditional traffic lights operate on predefined timers that do not consider real-time traffic conditions, resulting in unnecessary delays, increased fuel consumption, and higher carbon emissions. This paper proposes an Adaptive Traffic Signal Timer system using realtime computer vision and deep learning techniques to dynamically adjust traffic signal timings based on lane-wise vehicle density. The system employs the YOLO (You Only Look Once) object detection algorithm integrated with OpenCV to detect and count vehicles from live video streams. Based on calculated density metrics, optimal green signal durations are assigned dynamically. The proposed system improves traffic flow efficiency, reduces waiting time, and supports intelligent transportation and smart city infrastructure development.
Keywords: Intelligent Transportation System, Computer Vision, YOLO, Traffic Density Estimation, Adaptive Signal Control, Deep Learning.
Keywords: Intelligent Transportation System, Computer Vision, YOLO, Traffic Density Estimation, Adaptive Signal Control, Deep Learning.
Abstract
AI BASED LEGAL SIMPLIFICATION AND CASE OUTCOME PREDICTION
Prof. Dnyaneshwar Thombre, Mahesh Bhosale, Vikas Maurya, Intaza Chaudhary, Mausam Yadav
DOI: 10.17148/IJARCCE.2026.153112
Abstract: It's challenging for many people to understand the complexity of legal documents and legal process. Furthermore, most people do not know what type of legal knowledge they need to better understand legal policies, to forecast the outcome of a case, or to find a legal expert. This project is designed to develop an Artificial Intelligence (AI)- Powered Legal Document Simplifier and Legal Case Outcome Predictor. This project will assist people in understanding the legal document(s) related to their case and the potential outcome of the case.
The project will consist of three major parts. The first part will be the Legal Document Simplifier; using Natural Language Processing (NLP), this component will assist in simplifying legal documents. The second part will be the Case Outcome Predictor; utilizing machine learning algorithms (such as InLegalBERT) to provide a prediction of the outcome of a case (i.e., if the case will be approved or denied); and the rationale supporting the outcome. The final part will be the Lawyer Management System; through which individuals may submit their case data, and connect with a lawyer. This project will utilize AI, NLP, and web development technologies to develop a user-friendly interface that will enable the general public to better understand legal documents.
Keywords: Legal Document Simplification, Case Outcome Prediction, Natural Language Processing, InLegalBERT, Lawyer Management System.
The project will consist of three major parts. The first part will be the Legal Document Simplifier; using Natural Language Processing (NLP), this component will assist in simplifying legal documents. The second part will be the Case Outcome Predictor; utilizing machine learning algorithms (such as InLegalBERT) to provide a prediction of the outcome of a case (i.e., if the case will be approved or denied); and the rationale supporting the outcome. The final part will be the Lawyer Management System; through which individuals may submit their case data, and connect with a lawyer. This project will utilize AI, NLP, and web development technologies to develop a user-friendly interface that will enable the general public to better understand legal documents.
Keywords: Legal Document Simplification, Case Outcome Prediction, Natural Language Processing, InLegalBERT, Lawyer Management System.
Abstract
AUTOMATED DATA INTEGRATION AND REAL-TIME VISUALIZATION USING
NN AND POWER BI, D VIMAL KUMAR, A. VANITHA, MISHAL ALOY J
DOI: 10.17148/IJARCCE.2026.153113
Abstract: Nowadays, handling data by hand slows things down, causes more mistakes. Business intelligence needs faster ways to work, especially when updating dashboards, sharing alerts. One solution comes from using n8n, a free tool that links tasks together automatically. Instead of people managing each step, the system moves fresh data from Google Sheets straight into Power BI. Emails go out without anyone needing to click send. Updates happen on schedule, quietly, without delays. Testing showed clear gains - less time spent, fewer errors caught late. Information stays current because the flow never stops. Efficiency rises when repetitive jobs run themselves behind the scenes. Old methods fade once machines handle what humans used to do every day.
Keywords: One thing leads to another when tasks run on their own. Tools like n8n tie steps together without help.
Keywords: One thing leads to another when tasks run on their own. Tools like n8n tie steps together without help.
Abstract
SkillMirror: An AI-Based Communication and Interview Training Platform Using Speech and Computer Vision Analysis
Shreyas S. Gadekar, Atharv N. Mane, Prasad T. Kalavikattekar, Aniket M. Kamble, Karan P. Ghatage, Prof. Avani G Shahane
DOI: 10.17148/IJARCCE.2026.153114
Abstract: Effective communication and interview skills are essential for student employability, yet traditional training methods often lack personalization and real-time feedback. This paper presents SkillMirror, an AI-based communication and interview training platform that enhances both verbal and non-verbal skills through a multimodal analysis approach. The system simulates interview scenarios where user responses are captured via audio and video. Speech-to-text and natural language processing techniques are employed to evaluate grammar, fluency, and linguistic coherence, while computer vision methods analyze non-verbal cues such as facial expressions, posture, and eye contact. Based on these analyses, the platform provides real-time feedback, performance metrics, and personalized improvement suggestions.
Additionally, the system includes a quiz module for knowledge assessment and a feedback module for progress tracking and motivation. By integrating speech processing and behavioral analysis, SkillMirror offers a scalable and intelligent solution for improving communication skills and interview readiness.
Keywords: Artificial Intelligence, Interview Training, Speech-to-Text, Natural Language Processing, Computer Vision, Multimodal Analysis, Communication Skills, Real Time FeedbackVision, Multimodal Analysis, Communication Skills, Real-Time Feedback
Additionally, the system includes a quiz module for knowledge assessment and a feedback module for progress tracking and motivation. By integrating speech processing and behavioral analysis, SkillMirror offers a scalable and intelligent solution for improving communication skills and interview readiness.
Keywords: Artificial Intelligence, Interview Training, Speech-to-Text, Natural Language Processing, Computer Vision, Multimodal Analysis, Communication Skills, Real Time FeedbackVision, Multimodal Analysis, Communication Skills, Real-Time Feedback
Abstract
Human-Centered Machine Learning for Educational Systems: A Review of Prediction, Behavioral Analytics, and Explainable AI
Jayashree More*, Kirti Dinkar More, Sunita Nimba Deore
DOI: 10.17148/IJARCCE.2026.153115
Abstract: Machine learning (ML) has become a fundamental component of modern educational systems [1], [2]. Its applications extend from predictive analytics to learning analytics, personalization, and decision support [3]-[5]. This paper presents a comprehensive and human-centered review of ML applications in education using recent studies [1]- [10]. The study emphasizes interpretability, pedagogy, and ethical considerations. It identifies key application domains, commonly used techniques, and emerging research trends. Challenges such as bias, transparency, and scalability are critically examined [5], [10]. A structured literature review table and detailed discussion are provided. The paper concludes with research gaps and future directions for responsible and inclusive ML integration in education.
Keywords: Machine Learning in Education, Explainable AI, Learning Analytics, Personalized Learning, Educational Data Mining
Keywords: Machine Learning in Education, Explainable AI, Learning Analytics, Personalized Learning, Educational Data Mining
Abstract
Design and Development of a Role-Based Industry Management System Using Modern Web Technologies.
Mohd Raed, Arhan Shaikh, Abdul Gani Kazi, Ayan Shaikh
DOI: 10.17148/IJARCCE.2026.153116
Abstract: The rapid development of industrial operations has led to the need for effective management systems that can effectively manage different roles and resources in an organized manner. The traditional approach to managing industrial operations is usually characterized by a lack of proper management systems, leading to inefficiency and lack of transparency in the management process.
This paper focuses on the design and development of a web-based Industry Management System (IMS) that integrates different industrial management functions such as employee management and vendor management into a single platform. The proposed system is designed with a role-based approach that includes Industry Owner, Admin, Worker, and Vendor roles that provide a secure and organized approach to accessing the system functionalities.
The proposed Industry Management System is designed with the latest web technologies and provides a range of functionalities such as data management and real-time updates that improve the management process and communication between different roles and functions.
This research provides a clear example of the importance of a well-structured management system in the management of industrial operations and the need for a well-designed management system that can improve the management process and the level of productivity in the management process.
Keywords: Industry Management System, Next.js, React, Node.js, MongoDB, Role-Based Access Control, Web-based Application, Real-Time Communication.
This paper focuses on the design and development of a web-based Industry Management System (IMS) that integrates different industrial management functions such as employee management and vendor management into a single platform. The proposed system is designed with a role-based approach that includes Industry Owner, Admin, Worker, and Vendor roles that provide a secure and organized approach to accessing the system functionalities.
The proposed Industry Management System is designed with the latest web technologies and provides a range of functionalities such as data management and real-time updates that improve the management process and communication between different roles and functions.
This research provides a clear example of the importance of a well-structured management system in the management of industrial operations and the need for a well-designed management system that can improve the management process and the level of productivity in the management process.
Keywords: Industry Management System, Next.js, React, Node.js, MongoDB, Role-Based Access Control, Web-based Application, Real-Time Communication.
Abstract
Database-Driven Architecture for Academic Knowledge Modeling and Automated Practice Paper Generation
Haji Saad, Shaikh Mohammed Abbas, Tiwari Aditya, Naik Faiz, Shimpi Manasi
DOI: 10.17148/IJARCCE.2026.153117
Abstract: Students in diploma-level technical education often find it difficult to access academic content that is properly organized according to exam requirements. Most existing tools like study planners and question banks work separately and do not combine syllabus structure, previous year questions, and exam patterns in one place. This paper proposes a database architecture designed for managing academic content and supporting exam preparation. The system organizes syllabus elements such as subjects, units, topics, and intended learning outcomes using a structured relational model. It also includes previous year questions that are mapped to syllabus units and classified using Bloom’s taxonomy. In addition, the system represents exam patterns using slot-based structures and blueprint constraints. This helps in retrieving structured data and allows automatic generation of practice papers. Overall, the design shows how a well-organized database can support better academic understanding, question mapping, and exam focused preparation.
Keywords: Database-driven architecture, Academic knowledge modeling, Practice paper generation, Bloom’s taxonomy, Question mapping, Exam pattern modeling
Keywords: Database-driven architecture, Academic knowledge modeling, Practice paper generation, Bloom’s taxonomy, Question mapping, Exam pattern modeling
Abstract
An IoT-Driven Intrusion Detection and Autonomous UAV Surveillance System with Edge-AI Verification
Blessan B Kuriakose, Aswin Prakash, Krishnajith K S, Githu Manoj, Parvathy R Nair
DOI: 10.17148/IJARCCE.2026.153118
Abstract: Traditional home and commercial security systems relying solely on static CCTV cameras are highly vulnerable to physical tampering, often resulting in critical surveillance blind spots during targeted intrusions. Furthermore, conventional alarms suffer from high false-positive rates, leading to alarm fatigue. This paper proposes a robust, Internet of Things (IoT) driven, multi-layered security framework designed to provide uninterrupted monitoring through autonomous robotic response. At the core of the architecture, a localized edge-computing Ground Control Station (Laptop GCS) processes real-time video feeds, executing deep-learning algorithms for facial recognition and continuous pixel-variance analysis to mathematically detect sudden camera blackouts. In the event of a compromised camera, the system initiates a distributed, dual-tier response mechanism. Bypassing vulnerable local networks, the GCS publishes a low-latency execution payload to a secure HiveMQ Cloud MQTT broker. An Unmanned Aerial Vehicle (UAV), equipped with an onboard Raspberry Pi 5 companion computer and a Pixhawk flight controller, intercepts this payload, activates onboard hardware deterrents, and launches into an autonomous waypoint patrol, restoring visual oversight via a 5.8GHz analog video transmitter. Crucially, the system employs a secondary artificial intelligence verification loop; the GCS scans the incoming aerial feed for human silhouettes. Using adaptive decision-making, the system routes a high- priority alert and live video stream to a custom Flutter mobile application only upon visually confirming a human threat. Experimental results demonstrate a visual feed restoration time of under 14 seconds and near-zero false positives, providing a cost-effective, intelligent framework for next-generation smart security systems.
Keywords: Intrusion detection, Autonomous UAV, Edge Computing, MQTT, HOG algorithm, False-alarm mitigation, Smart home security.
Keywords: Intrusion detection, Autonomous UAV, Edge Computing, MQTT, HOG algorithm, False-alarm mitigation, Smart home security.
Abstract
Explainable Fake Job Posting Detection Using OCR, Machine Learning, and AI Reasoning
RAGUNATH M, Mrs. PRADEEPHA S, Dr E. MARIAPPAN, Dr M. KALIAPPAN
DOI: 10.17148/IJARCCE.2026.153119
Abstract: The increased availability of online recruitment sites has increased the possibility of fake job ads targeting potential employees. This paper proposes a transparent solution for the detection of fake job ads that utilizes Optical Character Recognition, a LRND ensemble classifier, SHAP, and an AI reasoning module. The proposed solution utilizes the EasyOCR library for Optical Character Recognition, TF-IDF for feature engineering, and a LRND ensemble classifier for the detection of fake job ads with an accuracy of 93.5%. SHAP is used for feature interpretation, and an AI reasoning module is used for providing explanations for the detection of fake job ads. The proposed solution is developed as a web application using the Flask framework.
Keywords: Fake job detection, Optical Character Recognition, TF-IDF, LRND classifier, SHAP, Flask, machine learning, fraud detection
Keywords: Fake job detection, Optical Character Recognition, TF-IDF, LRND classifier, SHAP, Flask, machine learning, fraud detection
Abstract
DATAVISION PRO: INTERACTIVE BUSINESS INTELLIGENCE DASHBOARD
Dr. Sajja Suneel, Manipatruni Trived, Nathi Dhana Sree
DOI: 10.17148/IJARCCE.2026.153120
Abstract: DataVision Pro, a web platform that integrates completely, was developed to eradicate the workflow gap between API testing and Business Intelligence (BI) visualization. The conventional approaches necessitate the usage of separate tools for confirming the responses of REST APIs and creating the reports thus causing inefficiency and delays in the team. DVP has come up with the solution to the problem by merging the main API testing and the data visualization characteristics plus the very good multi-user real-time collaboration and built-in BI tools all together. Some of the key features are real-time collaborative dashboard editing, automatic dashboard versioning, and secure sharing. The built-in BI features are creating KPIs with thresholds, blending data across charts, advanced drilldowns, and support for user- defined fields and time-series forecasting. The system is built on a serverless backend platform that guarantees real-time data synchronization for concurrent, intricate BI computations. Our research evaluates the efficiency of the platform, particularly its built-in BI feature integration effects on API testing workflows and visualization insights generation speed. We outline the system architecture and methods that provide an easy foundation for determining the performance and scalability of serverless, real-time BI collaboration. This indicates a substantial growth in developer productivity and the implementation of data supported decision making.
Index Terms: API testing, Business Intelligence, Real-time Collaboration, Data Visualization, Vertically scaled serverless computation, Data Mashup, Time Series Forecasting, Dashboard.
Index Terms: API testing, Business Intelligence, Real-time Collaboration, Data Visualization, Vertically scaled serverless computation, Data Mashup, Time Series Forecasting, Dashboard.
Abstract
Improved Customer Churn Estimation Using LSTM Networks
Dr. Sajja Suneel, Lokesh Gopal M, Harshitha P, Harshitha Reddy K
DOI: 10.17148/IJARCCE.2026.153121
Abstract: In the competitive landscape of financial services, the high cost of client acquisition makes retention a top strategic priority. This study addresses the limitations of conventional credit card churn prediction specifically rigid architectural constraints and inefficient categorical data processing by proposing a dynamic framework based on Long Short-Term Memory (LSTM) networks. By restructuring standard tabular datasets into”pseudo-sequences” and utilizing dense embeddings for high- cardinality features, we enable the model to capture nuanced temporal shifts in customer behavior. Our methodology evaluates four distinct LSTM configurations: Vanilla, Stacked, Bidirectional, and a hybrid Bidirectional-Stacked variant. By integrating these architectures into a unified ensemble, we achieved a peak classification accuracy of 92.35%. Beyond raw accuracy, the ensemble demonstrates exceptional recall performance. For banking institutions, this translates to a more reliable early- warning system that identifies at-risk accounts with precision, effectively reducing the revenue loss associated with undetected churn.
Index Terms: Customer attrition modeling, LSTM architectures, Deep learning frameworks, Ensemble modeling strategies, Credit risk analytics, Class imbalance handling
Index Terms: Customer attrition modeling, LSTM architectures, Deep learning frameworks, Ensemble modeling strategies, Credit risk analytics, Class imbalance handling
Abstract
Fireguard AI: Advanced Fire Detection and Evacuation Path Guidance System Using IoT and Artificial Intelligence
Harikrishnan S, Devika MS, Alex TJ, Vishnu PS, Ananthakrishnan CS, Ann Josy John
DOI: 10.17148/IJARCCE.2026.153122
Abstract: Fire accidents in buildings are a serious threat to human life and infrastructure, especially when the current fire evacuation systems in buildings are lack intelligent and adaptability. So, the solution proposes these system FireGurard AI. Traditional methods of detecting fire are using isolated sensors and alarm mechanism, which often delayed the response and inefficient evacuation. FireGuard AI integrate IOT sensors with deep-learning based computer vision (YOLO V8) for intelligent detection and evacuation planning. A Mobile application is developed to guide the evacuation safely during emergency. The proposed system detects the fire through the CCTV video streams of the building by a YOLO model, then the risk level is evacuated by physical agent which is integrated with IoT sensors and a camera module. The sensor data and visual analysis is to fixed to assess risk level and guide the occupant through safe evacuation path by mobile app and physical actuation like voice assistance, and lights. Using adaptive decision making large experimental evacuation shows that the system improve detection accuracy, reduce false alarm and enhance evacuation efficiency compared to traditional approaches. This proposed solution provides cost effective and intelligent framework for next generation fire system for smart buildings.
Keywords: Fire detection, YOLO V8, Emergency guidance, Path Planning, Evacuation planning, A* path finding algorithm.
Keywords: Fire detection, YOLO V8, Emergency guidance, Path Planning, Evacuation planning, A* path finding algorithm.
Abstract
Lok Netra: A Vision of Politics & Voting Power
Veer Panchal, Chinmay Thukarul, Shubham Toraskar, Mrs. Nisha Karolia
DOI: 10.17148/IJARCCE.2026.153123
Abstract: Democratic systems rely on informed citizen participation; however, the rapid spread of digital misinformation and limitations of traditional voting methods reduce both awareness and accessibility. This paper presents Lok Netra, a web-based civic engagement platform that integrates political awareness, analytics, and secure remote e-voting within a unified ecosystem.
The system leverages a serverless architecture using Firebase, enabling scalable data handling without traditional backend constraints. Real-time political news is aggregated through GDELT APIs, while candidate information is dynamically retrieved using Wikipedia services. A client-side biometric authentication module, powered by face- api.js, ensures secure voter verification using facial recognition and liveness detection.
Additionally, Lok Netra provides historical election analytics (1962–2024) and real-time predictive visualizations to enhance transparency. Built with React (Vite) and Tailwind CSS, the platform demonstrates how modern web technologies can deliver a secure, scalable, and user-centric solution for digital democracy.
Keywords: React Web Applications, Serverless Architecture, E-Governance, Political Awareness, Facial Recognition, Digital Democracy, Cloud Firestore, Biometric Authentication, Civic Tech, Data Analytics.
The system leverages a serverless architecture using Firebase, enabling scalable data handling without traditional backend constraints. Real-time political news is aggregated through GDELT APIs, while candidate information is dynamically retrieved using Wikipedia services. A client-side biometric authentication module, powered by face- api.js, ensures secure voter verification using facial recognition and liveness detection.
Additionally, Lok Netra provides historical election analytics (1962–2024) and real-time predictive visualizations to enhance transparency. Built with React (Vite) and Tailwind CSS, the platform demonstrates how modern web technologies can deliver a secure, scalable, and user-centric solution for digital democracy.
Keywords: React Web Applications, Serverless Architecture, E-Governance, Political Awareness, Facial Recognition, Digital Democracy, Cloud Firestore, Biometric Authentication, Civic Tech, Data Analytics.
Abstract
AI AND IOT-ENABLED AUTONOMOUS ANTI-DRONE DEFENSE SYSTEM
Aien Joel Paul, Akhil Ginoy, Albin Babu, Mahesh T.S, Ann Josy John
DOI: 10.17148/IJARCCE.2026.153124
Abstract: The rapid growth of unmanned aerial vehicles (UAVs), commonly known as drones, has created new opportunities across industries such as surveillance, agriculture, and delivery systems. However, unauthorized or malicious drone usage poses serious security risks to sensitive areas including airports, military installations, and public infrastructure. This paper presents the design and implementation of an autonomous drone detection, tracking, and countermeasure system using an edge computing platform. The proposed system utilizes a deep learning-based object detection model implemented through YOLOv8 to identify drones from real-time video captured by a camera module. The system is deployed on a compact embedded platform based on Raspberry Pi 5 and optimized using the NCNN inference framework to enable efficient edge AI processing. Once a drone is detected, a pan–tilt servo mechanism automatically tracks the object using real-time positional feedback. The system also integrates a directional countermeasure module based on ESP32 and NRF24L01 to demonstrate a 2.4 GHz signal interference mechanism. A web-based monitoring interface provides live video streaming through HTTP Live Streaming generated using FFmpeg, enabling remote system monitoring and event logging. Experimental evaluation demonstrates the feasibility of implementing an intelligent drone monitoring system using low- cost embedded hardware and open-source software technologies.
Keywords: Drone Detection, UAV Monitoring, Edge AI, YOLOv8, NCNN, Raspberry Pi, Computer Vision, Drone Countermeasure, HLS Streaming, Embedded Systems
Keywords: Drone Detection, UAV Monitoring, Edge AI, YOLOv8, NCNN, Raspberry Pi, Computer Vision, Drone Countermeasure, HLS Streaming, Embedded Systems
Abstract
MULTI-LAYERED AUTHENTICATION AND ENCRYPTION FOR CLOUD SAFETY
P V Ramana Murthy, Mohammad Noman Malik, Gampa Rahul, Talari Ajay Kumar, Gurram Ravi Kiran
DOI: 10.17148/IJARCCE.2026.153125
Abstract: This paper introduces Cloud Safety, a robust multi-factor authentication (MFA) framework that integrates password-based authentication, Time-based One-Time Password (TOTP), and SMS OTP to enhance identity assurance in cloud environments. The framework addresses critical vulnerabilities in traditional single-factor systems—such as phishing, brute force, and replay attacks—by implementing layered security controls without compromising usability. Additionally, encryption mechanisms are applied at the data layer to protect stored information. Designed for real-world deployment, Cloud Safety emphasizes low-latency performance, scalability, and adaptability to modern threat models. Comparative analysis and implementation results demonstrate its effectiveness in mitigating common cloud security threats while maintaining a practical balance between security and user experience. Future enhancements may include biometric integration and adaptive, AI-driven access control.
Keywords: cloud authentication; multi-factor authentication; authentication factors; cloud intrusion detection; user behavior
Keywords: cloud authentication; multi-factor authentication; authentication factors; cloud intrusion detection; user behavior
Abstract
GraphSense-RAG: An Intelligent Multi-Modal Document Question Answering System
Dr. M. Purnachandra Rao, D. Mastan Bee, B. Navya, B. Sravani, A. Bhargavi, B. Uma
DOI: 10.17148/IJARCCE.2026.153126
Abstract: The rapid growth in the digital documents made a critical challenge in retrieving accurate information from the large volumes of data. This question-answering system has become a challenging task for the companies and individuals who deals with the huge volume of information. To overcome these challenges, the proposed mainly focuses on Document Question Answering system based on Retrieval-Augmented Generation approach which is integrated with the embedding search. The main aim of this project is to provide accurate and reliable information for the user query from the document. The proposed system supports multi modal data processing and also uses hybrid retrieval technique which includes semantic search, keyword matching and metadata filtering. These techniques are used to enhance the retrieval performance and accuracy. The proposed system also included with a dedicated hallucination detection mechanism which validates the generated result by grounding them in retrieved evidence. In addition to this the system also supports automatic document summarization and indexing. It also allows the user for efficient ingestion of new documents. The proposed system provides scalable, reliable solution which can be effectively used for academic research and also for the enterprise companies.
Index Terms: Retrieval-Augmented Generation, Vector Embedding Search, Multi-Modal Question Answering, Hybrid In- formation Retrieval, Knowledge Graph Integration
Index Terms: Retrieval-Augmented Generation, Vector Embedding Search, Multi-Modal Question Answering, Hybrid In- formation Retrieval, Knowledge Graph Integration
Abstract
AI Career Role Recommendation System with Skill Gap Analysis, Career Progression, and Salary Simulation Using Resume Semantic Analysis
Ms. Neelam Mary Vijaya Nirmala, G. Pallavi, B. Jhansi Lakshmi, G. Ramya, J. Divya Dimple
DOI: 10.17148/IJARCCE.2026.153127
Abstract: Due to changing job market trends, differing skill sets, and the absence of personalized career counselling and guidance systems, it has become difficult for job seekers or students to choose their path. Therefore, an AI-Based Career Guidance and Role Recommendation System is proposed in this project, which performs smart resume semantic analysis to recommend career roles based on skills, educational qualifications, amount of experience, and professional interests. Resume parsing and Natural Language Processing (NLP) based techniques are used to convert tacit knowledge contained in a candidate’s resume, which is an unstructured collection of words, into structured information about technical skills, educational qualifications, professional certifications, and work experience. TF-IDF (Term Frequency – Inverse Document Frequency) vectorization is used to convert a resume into a high-dimensional feature vector. Cosine similarity based semantic similarity algorithms are then applied to map these candidate’s profiles to role descriptor templates and industry-specific competency frameworks of matching job roles. The system includes a Skill Gap Severity Index to identify missing and underserved skills for target roles, a Customized Learning Plan Generator to recommend upskilling pathways, a Career Progression Simulation to visualize different career paths, and a Salary Range and salary levels. Peer Benchmarking System compares a users and Resume Improve intelligence improves resumes by analyzing them to identify missing skills and errors instead of only giving a score. profile with similar candidates having the same degree and experience level. The proposed system is based upon semantic profiling, similarity modeling, predictive career analytics, and resume improvement intelligence, providing students, freshers, and job seekers with personalized, accurate, and actionable recommendations to make informed career decisions and achieve sustainable growth in their respective careers.
Index Terms: Career Role Recommendation, Resume parsing, Skill Extraction, TF-IDF, Semantic Similarity, Artificial Intelligence, Salary Range & Simulation, Natural Language Processing (NLP), Skill Gao Severity Index, Skill-Based Career Mapping.
Index Terms: Career Role Recommendation, Resume parsing, Skill Extraction, TF-IDF, Semantic Similarity, Artificial Intelligence, Salary Range & Simulation, Natural Language Processing (NLP), Skill Gao Severity Index, Skill-Based Career Mapping.
Abstract
Campus Placement Eligibility & Recruitment Outcome Prediction Using Academic and Behavioural Attributes
Mrs. M. Khamar, D. Sripriya, B. Vyshnavi, D. Gowthami, G. Sushma
DOI: 10.17148/IJARCCE.2026.153128
Abstract: Campus placement outcomes influence both students career and also the academic situation of education institutions. In the recent years, employment market has become very competitive and difficult for training and placement departments to assess the students accurately for the placements using traditional evaluation methods. To address this challenge, this system is presenting a machine learning-driven framework for predicting the campus placements and its eligibility by evaluating academic and behavioural attributes. This proposed approach uses features such as cumulative grade point average (CGPA), students technical skills, internship experience, and also communication capabilities to estimate placement probability. The system follows a structured pipeline that includes data collection, data preprocessing, feature selection, train test split, machine learning models, model evaluation and Placement predicted outcome. Data preprocessing techniques can be used to reduce this. Correct inconsistencies, encode categorical variables, and normalize numerical features, enabling the models to be built. Thus, a number of different classification algorithms are implemented and tested to see which works best Prediction. The framework is implemented in the python programming language and the pandas library is used. And analysis, and Scikit-learn can be used for modelling and evaluating the approach. Experimental Analysis shows that combining both academic and behavioural data can yield better results. Accuracy prediction as compared to models that only use the academic performance for prediction. The result experimental and communication-related features play an important role in improving employability and placements preparation. The proposed system provides actionable insights that can help training and placement departments to help identify students that need direct skill instruction and to target those needing specific skills programs or grouping like skilled students together for training. By leveraging machine learning and data- driven decision making, this research contributes to improving placement preparation strategies and optimizes institutional placement outcomes.
Index Terms: campus placement eligibility, python, pandas, Scikit-learn, classification algorithms.
Index Terms: campus placement eligibility, python, pandas, Scikit-learn, classification algorithms.
Abstract
Multi Model Emotion Aware Conventional Chatbot Using Facial Expression and Text Sentiment Fusion
Mrs. K. Tejaswi, Ch. Pushpa Manasa, D. Hima Sravanthi, B. Pravallika, G. Girishma
DOI: 10.17148/IJARCCE.2026.153129
Abstract: Understanding or analyzing the human emotion is difficult for smart, adaptive human-computer interaction systems. This project proposed a multi-modal emotion-aware conversational chatbot that recognizes emotions from the users by the help of facial expression, text-based sentiment analysis from speech-to-text conversion and also written text. Facial emotions are extracted using computer vision techniques. Such as eye movement, mouth shape, and eyebrow position, based on all these it captures the emotion state. while speech is converted to text, and also by direct text using natural language processing (NLP) methods. The emotional information obtained from both modalities is combined using a weighted fusion mechanism to clarify the user’s overall emotional state. Based on this, the chatbot generates emotionally appropriate and relevant context- aware responses. The proposed system adapts human-centered interactions and shows its utility for mental health support. Using OpenCV facial expression recognition, VGG16 convolutional neural networks, automatic speech recognition (speech-to-text), VADER sentiment analysis, and weighted multimodal fusion. Experimental results show that the performance of emotion recognition information response relevance compared to single- modality systems, leading to the that multimodal emotion fusion greatly improves the effectiveness and empathy of conversational AI systems.
Index Terms: Multimodal Emotion Analysis, Emotion-Aware Chatbot, Facial Expression Recognition, Sentiment Analysis, OpenCV, VGG16, Natural Language Processing, VADER
Index Terms: Multimodal Emotion Analysis, Emotion-Aware Chatbot, Facial Expression Recognition, Sentiment Analysis, OpenCV, VGG16, Natural Language Processing, VADER
Abstract
Virtual Personal Productivity Assistant with Task Automation and Intelligent Scheduling Capabilities
Mr. P. Madhubabu, Ch. Maheswari, A. Thanusha, B. Nikitha, Ch. Susmitha, Ch. Lavanya Chowdary
DOI: 10.17148/IJARCCE.2026.153130
Abstract: The increasing amount of digital work has made it more difficult to manage daily tasks, appointments, and academic obligations. To help people, to organize their tasks and to maximize time management through automation a Virtual Per- sonal Productivity Assistant was developed. Python-built assistant has provided few features like Task planning, alerts, calendar integration, and basic workflow automation. one of the assistant’s primary functions is identifying tasks from informal texts such as emails and chat messages and converting them in to to-do items. In order to improve focus, the system also plays ambient sounds and productivity music tailored to each task. Moreover, this system considers the user’s mood and energy levels as inputs to prioritize task and timing. By utilizing natural language processing technologies user can interact with the system through voice or text commands. Ultimately, this project is a practical tool for improving productivity by reducing manual efforts and prompting efficient time management.
Index Terms: Virtual Personal Productivity Assistant, Task Management, Automation, Python, Natural Language Processing.
Index Terms: Virtual Personal Productivity Assistant, Task Management, Automation, Python, Natural Language Processing.
Abstract
Conversational AI Travel Assistant for Automated Trip Planning and Destination Query Interpretation
Mrs. V. Krishna Vijaya, J. Hemanth Kumar, J. Mallikarjuna Rao, G. Veera Narayana, D. Siva Manikanta, A. Pratap
DOI: 10.17148/IJARCCE.2026.153131
Abstract: Now a days tourism industry is gradually increasing, traveller’s are facing issues in planning their trips, the old planning methods are often causing decision making problems due to hours of our research in inconsistent platforms for budgets, destinations and attractions. To overcome these challenges, our project introduces a Conversational AI Travel Assistant that provides the journey planning through the natural and human being-like interconnection by acts as an intelligent digital assistant. It uses the Natural Language Processing (NLP) technology, our system fulfils the user needs such as destination preferences, budget flexibility, traveling time, travel dates, and seasonal attractions. It also extracts the key requirements and creates highly customized advice along with the detailed plan, day-wise schedule. Our AI Travel Assistant system is developed by using Python as a web-based application, it offers high availability and an automatic interface while providing context-aware and suitable responses that constantly decrease the time and effort of the user in trip organization. This project provides the actual application of artificial intelligence in the tourism industry like real-time travel alerts, voice-based interaction, and weather forecast.
Keywords: Conversational AI, Natural Language Processing, Travel Planning, Intent Recognition, Itinerary Generation, Tourism Technology
Keywords: Conversational AI, Natural Language Processing, Travel Planning, Intent Recognition, Itinerary Generation, Tourism Technology
Abstract
AI Driven Personalized Diet and Nutrition Recommendation System Based on Health Metrics and Food Knowledge Graphs
Mrs. J. Mounika, E. Madhan Kumar, A. Santhosh, B. Salem Raju, D. Veera Brahmam
DOI: 10.17148/IJARCCE.2026.153132
Abstract: Maintaining a healthy and balanced diet is not an easy task because everyone has different health needs, daily routines, budgets, and eating habits.Most current diet recommendation systems use fixed rules or common food guidelines.These systems often do not consider important personal health details like allergies, daily activity, or calorie requirements.They also ignore when meals should be eaten and whether the food is low-cost, which makes these systems less practical. Because of this, many people find it hard to follow the diet plans they suggest. To solve these problems, this research presents an AI-based system that suggests personalized diet and nutrition plans for each user. The system analyzes user information such as BMI, daily activity, health conditions, allergies, and calorie requirements to create customized meal plans.It uses intelligent techniques to suggest the right type of diet for each person and a food knowledge graph to understand how foods are linked to nutrients and dietary rules. The system also recommends meals based on the time of day and ranks them to ensure proper nutrition and calorie balance.The system also suggests low-cost meals and replaces expensive ingredients with cheaper alternatives. Overall, it gives simple, practical, and personalized diet suggestions that help people eat healthier.
Keywords: Python, ML Classifiers, Nutrition Dataset, Recommendation Engine
Keywords: Python, ML Classifiers, Nutrition Dataset, Recommendation Engine
Abstract
RASA-Based End-to-End Conversational Chatbot with Intent Training and Dialogue Policy Management
Mrs. B. Kalyani, A. Datta Maheswara Das, D. Yogeswar, G. Ajay, D. Ramakrishna, G. SreeRam
DOI: 10.17148/IJARCCE.2026.153133
Abstract: The RASA-Based End-to-End Conversational Chat- bot with Intent Training and Dialogue Policy Management is an intelligent and scalable conversational AI system designed to automate user interactions across various domains such as customer support, education, healthcare information services, and business automation. This project focuses on developing a chatbot using the RASA framework and Python, enabling the system to understand user inputs through Natural Language Understanding (NLU) techniques including intent classification and entity extraction. NLU pipelines are trained using domain- specific datasets to accurately interpret diverse user queries expressed in natural language. Dialogue Policy Management is implemented using RASA stories and rules to maintain con- versation context and manage multi-turn interactions effectively, allowing the chatbot to respond dynamically rather than relying on static question–answer mechanisms. By reducing manual in- tervention and improving response accuracy and consistency, the chatbot enhances operational efficiency and user experience. This project demonstrates the practical application of conversational AI concepts such as intent training, dialogue state tracking, and policy optimization, highlighting the effectiveness of RASA as an open-source framework for building robust, real-world conversational systems.
Keywords: RASA, Python, NLU Pipelines, Dialogue Policies
Keywords: RASA, Python, NLU Pipelines, Dialogue Policies
Abstract
Healthcare Consultation and Symptom Response Chatbot Powered by Medical NLP Dialogue Understanding
Mrs. K. Deepthi, A. Ratna Sai, CH. Karthik, G. Hemanth Kumar, G. Venkata Sai Teja
DOI: 10.17148/IJARCCE.2026.153134
Abstract: People can get information and services from chat- bots in a way that feels like talking to someone. Medical care is very important to everyone. It can be hard to go see a doctor for everything that is wrong with us. We want to make a healthcare chatbot that uses intelligence and can give people some basic information about what might be wrong with them before they go see a doctor. The chatbot uses a kind of technology called Natural Language Processing. This technology helps the chatbot understand what people are saying in their words. This makes it easy for people to talk to the chatbot and get help. The chatbot can look at what symptoms a person’s experiencing and suggest some possible treatments, for medical issues. People need help with care and that is what the healthcare chatbot is, for. The healthcare chatbot is a tool that helps people get the care they need. It looks at the symptoms to figure out if the problem’s serious or not. The user should still go see a doctor. If the healthcare chatbot says the problem is not serious the user can take care of it themselves. If the healthcare chatbot says the illness is serious the user should go see a doctor because the healthcare chatbot is not a doctor. Our system looks at the symptoms of the case to help the user pick the medicine and things to prevent the illness. The Python package Natural Language Toolkit does a job of understanding English language that is written in code. It looks at what the user says and gives answers that sound like they are, from a person. The Natural Language Toolkit is used for Natural Language Processing. It helps the chatbot understand what the user is saying and gives answers. The chatbot uses Natural Language Toolkit to make sure the user gets an answer.
Abstract
Dynamic AI Storytelling and Narrative Plot Generation Engine Using Generative Language Models
Mr. P. Madhubabu, B. Lakshmi Srinivas, B. Pavan Kumar, B. Prabhu, A. Nehemya
DOI: 10.17148/IJARCCE.2026.153135
Abstract: The proposed system mainly focuses on generating Dynamic AI storytelling and also Narrative plot Generation by using models like generative language models. The main aim of this proposed system is to create an interactive story which is changing according to the emotions and choices of user. The traditional systems mainly focus on linear or semi-linear story creation whereas this proposed system introduces dynamic branching which means allowing the user to change the story dynamically. This system also provides the story to branch into multiple according to the perspectives of the user. Additionally, this system also identifies the emotional tones of the user like anger, happiness, sadness, anxiety etc. Based on these emotional tones of the user this system generates the next part of the story. Proposed system generates the characters as a realistic and these characters also stores the past actions that are given by the user choices. System provides the consistent and realistic characteristics in the story that is generated by system. This system changes the story in different ways according to the user preferences. This system generates more interactive and creative stories, where these stories are meaningful and remains logical. AI is used to generate dynamic stories whereas NLP is used to help the system to understand the language, emotions and context of the user and also makes the story to be natural.
Keywords: Python, GPT Models, Transformers, Text Generation APIs
Keywords: Python, GPT Models, Transformers, Text Generation APIs
Abstract
Real-Time Object Detection System Using YOLO-Based Vision Models
Mrs. K. Deepthi, B. Akash, Revanth Kumar, Ch. Tharun, K. Narendra
DOI: 10.17148/IJARCCE.2026.153136
Abstract: Object detection plays an important role in many real-time applications such as surveillance, traffic monitoring, smart automation. In recent years, deep learning techniques have significantly improved the accuracy and speed of object detection systems. This project presents the implementation of a real-time object detection system. Using YOLO integrated with CNN. The proposed system processes video frames, extracts features using CNN layers, and detects objects. By predicting bounding boxes and class labels in a single step. YOLO enables fast detection while maintaining acceptable accuracy, making it suitable for real-time environments. Experimental results show that the system performs efficiently on live video streams with good accuracy and real-time speed. Proving its suitability for practical applications.
Keywords: Real-time object detection, YOLOv8, CNN, Python, OpenCV, Smart monitoring.
Keywords: Real-time object detection, YOLOv8, CNN, Python, OpenCV, Smart monitoring.
Abstract
A Smart Cybersecurity Awareness Website for User Education via Courses and Gamified Learning
Rakhangi Mohammad Saif, Maniyar Mohammad, Shaikh Shahim, Shaikh Rehan, Shaikh Shahista
DOI: 10.17148/IJARCCE.2026.153137
Abstract: Users in today’s digital environment often struggle to understand and apply cybersecurity concepts due to a lack of structured and engaging learning resources. Most existing cybersecurity awareness tools and training platforms provide either theoretical content or isolated quizzes, but they do not effectively combine structured learning, practical assessment, and user engagement in a single system. This paper proposes a web- based cybersecurity awareness platform that utilizes gamification to improve user engagement and learning outcomes. The system organizes cybersecurity topics such as phishing, malware, password security, and safe browsing practices into structured modules with multiple levels. It also includes interactive quizzes, real-time feedback, and scenario-based activities mapped to different difficulty levels. In addition, the system incorporates gamification elements such as points, badges, and leaderboards. This helps in improving user participation and allows continuous performance tracking. Overall, the design demonstrates how an integrated and interactive platform can support better cybersecurity understanding, engagement, and safe online practices.
Keywords: Cybersecurity awareness, Gamification-based learning, Web-based learning platform, Interactive quizzes, User engagement, Phishing detection, Malware awareness, Performance tracking, Scenario-based learning, Digital safety practices
Keywords: Cybersecurity awareness, Gamification-based learning, Web-based learning platform, Interactive quizzes, User engagement, Phishing detection, Malware awareness, Performance tracking, Scenario-based learning, Digital safety practices
Abstract
Smart Budget Buddy: An AI-Enhanced Personal Finance Management System Using Large Language Models and Serverless Architecture
Khan Ayaan, Shaikh Rehan, Malik Aban, Ansari Rayyan, Neha Salunkhe
DOI: 10.17148/IJARCCE.2026.153138
Abstract: This paper is about Smart Budget Buddy, a financial management tool that uses artificial intelligence to make it easy for people to track their expenses see where their money is going and get advice on how to manage their finances. The system is made up of a web application that works with a serverless backend, which uses PostgreSQL to keep user data safe and secure. One of the things about Smart Budget Buddy is that it uses a big language model called Gemini to automatically sort transactions and give users personalized advice on their finances all without sharing sensitive information with the users device. The goal of Smart Budget Buddy is to be secure work well for a lot of users and make it easy for people to make financial decisions without having to do a lot of work. When we tested Smart Budget Buddy we found that it was really good, at sorting transactions and giving advice, which shows that using big language models can be a great way to build personal finance tools that work well and are easy to use.
Keywords: —Artificial Intelligence, Personal Finance, Large Language Models, Serverless Architecture, React, Supabase, Gemini, Budget Management, Edge Computing, TypeScript, PostgreSQL, Row Level Security.
Keywords: —Artificial Intelligence, Personal Finance, Large Language Models, Serverless Architecture, React, Supabase, Gemini, Budget Management, Edge Computing, TypeScript, PostgreSQL, Row Level Security.
Abstract
Crime Report and Reward Management System Using Machine Learning
Meghana Kawale, Vrushali Nikam, Shivani Sagare, Madiha Mujawar, Tejaswini Khot, R. S. Kamble
DOI: 10.17148/IJARCCE.2026.153139
Abstract: This paper presents an AI-Driven Crime Reporting and Reward Management System to improve crime reporting efficiency and transparency. The system allows citizens to report crimes digitally with optional anonymity, while Machine Learning techniques are used for crime classification, fake report detection, and hotspot analysis. A reward mechanism encourages responsible reporting by evaluating the accuracy of submitted information. The system enhances decision-making, reduces response time, and strengthens trust between citizens and law enforcement.
Keywords: Artificial Intelligence, Crime Reporting, Machine Learning, Reward System, Crime Analysis
Abstract
AI CCTV ACCIDENT DETECTION
SUMESH NK, REVATHI A
DOI: 10.17148/IJARCCE.2026.153140
Abstract: Road accidents are one of the major causes of injuries and fatalities around the world. In many cases, delays in identifying accidents and informing emergency services increase the severity of the situation. This project focuses on developing a system that can detect road accidents using CCTV cameras. and artificial intelligence techniques to automatically detect road accidents the system analyses video footage captured from surveillance cameras to identify vehicle movement and possible accidents. YOLO (You Only Look Once).
When an abnormal situation such as a collision or sudden stop is detected, the system determines the location of the accident using a grid-based mapping method and sends an alert to the nearest patrol unit via email. The proposed system is developed using Python and computer vision libraries. This approach helps in reducing the response time of emergency services and improving road safety.
Keywords: Artificial Intelligence, CCTV, Accident Detection, YOLO, Smart Surveillance
When an abnormal situation such as a collision or sudden stop is detected, the system determines the location of the accident using a grid-based mapping method and sends an alert to the nearest patrol unit via email. The proposed system is developed using Python and computer vision libraries. This approach helps in reducing the response time of emergency services and improving road safety.
Keywords: Artificial Intelligence, CCTV, Accident Detection, YOLO, Smart Surveillance
Abstract
BLOCK-CHAIN BASED DOCUMENT VERIFICATION SYSTEM USING IPFS
Prof. Swapna V. Tikore, Akash Devade, Vyankatesh Kulkarni, Sandip Pawar, Ashwin Ingle
DOI: 10.17148/IJARCCE.2026.153141
Abstract: In this project, we proposed a blockchain-based solution and framework for document sharing and version control to facilitate multi-user collaboration and track changes in a trusted, secure, and decentralized manner, with no involvement of a centralized trusted entity or third party. This solution is based on utilizing Ethereum smart contracts to govern and regulate the document version control functions among the creators and developers of the document and its validators. Moreover, our solution leverages the benefits of IPFS (InterPlanetary File System) to store documents on a decentralized file system. The proposed solution automates necessary interactions among multiple actors comprising developers and approvers. Smart contracts have been developed using Solidity language, and their functionalities were tested using the Remix IDE (Integrated Development Environment). The paper demonstrates that our smart contract code is free of commonly known security vulnerabilities and attacks.
Keywords: Blockchain, IPFS, Document Verification, Ethereum, Smart Contracts, Decentralized Storage.
Abstract
aLLoyM: “Phase Diagram Prediction System”
Sakshi K. Kamble, Anagha G. Harshe, Soniya C. Dhupdale, Shubham U. Dharwat, Soham P. Kapileshwar
DOI: 10.17148/IJARCCE.2026.153142
Abstract: Phase diagrams are essential thermodynamic
tools that describe equilibrium phase stability as functions of temperature and
composition in alloy systems. They guide alloy design, heat-treatment
optimization, and microstructural control in critical industries. However,
experimental determination of phase diagrams is costly and time-intensive.
Although computational approaches such as CALPHAD (Calculation of Phase
Diagrams) provide reliable thermodynamic modelling through Gibbs free energy
minimization, they depend on curated parameter databases and expert assessment,
limiting rapid exploration of new material systems.
In this work, we introduce aLLoyM, a
domain-adapted Large Language Model (LLM) developed for structured alloy phase
diagram prediction. Thermodynamic data generated from CALPHAD assessments in
the Computational Phase Diagram Database (CPDDB), covering 389 binary and 38
ternary systems, were systematically sampled to produce over 800,000
equilibrium data points. These data were transformed into multi-task
Question–Answer (Q&A) pairs and used to finetune the Mistral-Nemo-Instruct
model via Low-Rank Adaptation (LoRA), enabling efficient domain specialization.
The framework supports three thermodynamic
reasoning tasks: full phase information prediction, phase name inference, and
inverse experimental condition prediction. Performance was evaluated under both
interpolation and extrapolation settings to assess generalization. Results show
substantial improvement over baseline LLM performance and demonstrate the
model’s ability to infer plausible phase behaviour for previously unseen
systems.
These findings highlight the potential of
integrating
Large Language Models with computational
thermodynamics to develop scalable AI-assisted tools for accelerating alloy
design and materials discovery.
Abstract
MULTI-LEVEL SIGN LANGUAGE RECOGNITION SYSTEM
NALINA P, Dr REVATHI A
DOI: 10.17148/IJARCCE.2026.153143
Abstract: Sign language is an important medium of communication for hearing and speech-impaired people. Nevertheless, there still exist communication gaps between sign language speakers and the general public owing to a lack of understanding of sign language gestures. This paper presents a Multi-Level Indian Sign Language (ISL) Recognition System that can identify word-level as well as sentence-level gestures from video inputs. The proposed system employs a hybrid deep learning model that combines Convolutional Neural Networks (CNN) for spatial feature extraction, Bi- Directional Long Short-Term Memory (BiLSTM) for modeling temporal sequences, and an Attention Mechanism to selectively, concentrate on important frames in the gesture sequence. The system analyzes video frames, extracts spatial- temporal features, andclassifiesthem to predict the corresponding word and sentence. Besides text output, the system also offers multimodal feedback in terms of synthesized speech and visualization of output, thus improving accessibility and usability. The proposed method is expected to fill the communication system.
Keywords: Indian sign language (ISL),CNN, BiLSTM, Attention mechanism, video gesture recognition, multimodal assistive system.
Keywords: Indian sign language (ISL),CNN, BiLSTM, Attention mechanism, video gesture recognition, multimodal assistive system.
Abstract
Leveraging Machine Learning for Intelligent Financial Forecasting and Investment Decision Support
Mrs. Dhanashri Kulkarni, Siddhi S. Shilahar, Pushkar D. Kaslikar, Pratik Pradip Kale, Chaitanya V. Kaypure, Parth P.Kshirsagar
DOI: 10.17148/IJARCCE.2026.153144
Abstract: In today’s rapidly evolving financial world, predicting market behaviorhas become more challenging than ever. Investors and financial institutions must deal with constantly changing trends, large volumes of data, and uncertain economic conditions, making traditional forecasting techniques increasingly insufficient. Conventional statistical methods often fail to capture the complex, non-linear nature of financial markets, resulting in delayed responses and unreliable predictions. To address these limitations, this paper explores the use of Machine Learning (ML) to develop a smarter and more reliable financial forecasting system. The proposed framework analyzes historical market data, identifies meaningful patterns, and uses predictive intelligence to estimate future price movements. Deep learning models, particularly Long Short-Term Memory (LSTM) networks, are employed to effectively handle time-series data and improve predictive performance. Initial findings suggest that AI-driven forecasting offers better accuracy and faster insights compared to traditional approaches. By supporting timely and informed decision-making, this AI-based financial system encourages a shift from reactive investment strategies to proactive, data-driven planning, ultimately helping to reduce risk and improve overall financial outcomes.
Abstract
AI Powered Low Code/No Code Software Development
Dilip Vishwakarma, Anil Vasoya, Aruna Pawate
DOI: 10.17148/IJARCCE.2026.153145
Abstract: The Low-Code and No-Code (LCNC) platforms haverecently been recognized as a strategic move to speed up application development, minimize the need forspecialized programming knowledge, and quicklychange the digital environment of the enterprise.The current literature has consistently shown theproductivity benefits of LCNC platforms throughabstraction, visual modeling, and component-basedreuse; however, these benefits have been shown to be applicable only to standardized and simpleapplications. With the increasing use of LCNC platformsfor developing large- scale enterprise applications,the current issues of scalability, architecturalrobustness, quality, governance, and security havebecome more visible. Recently, the integration ofArtificial Intelligence (AI) into LCNC platforms hasreceived increasing attention, and the benefits ofintelligent automation, decision support, and adaptiveworkflow design have been widely explored.However, the new agency and reasoning paradigmsof AI have introduced new issues of explainability,compliance, and controlled autonomy in end-userdevelopment. This paper offers a thorough literature survey and comparative synthesis of existing research on LCNC platforms, specifically with regard to productivity results, drivers of adoption, architectural strategies, and AI integration. The results of this analysis indicate that existing research is still piecemeal and lacks empirical support for architectural patterns, quality, and AI-integrated workflows in an enterprise setting. Moreover, although AI-supported low-code development appears promising, empirical data on the productivity effect, governance aspects, and return on investment of AI-assisted development is still relatively scarce. On the basis of the synthesized results, this paper points out the key research gaps and open challenges in terms of scalability, architectural validation, governance, security, and agentic AI integration. The findings of this research offer a systematic basis for future research work on designing scalable, secure, and intelligent AI-assisted LCNC platforms for sustainable enterprise adoption.
Keywords: Low-Code/No-Code, AI-assisted software engineering, architectural governance, testability, enterprise software, developer productivity.
Keywords: Low-Code/No-Code, AI-assisted software engineering, architectural governance, testability, enterprise software, developer productivity.
Abstract
A Study Impact on the Influencer Marketing on Brand Trust and Purchase Intention: A Comparative Study of Millennials andGenZinIndia
Shalin Dwivedi, Ishita BhanushaliandDr. Hiren Harsora
DOI: 10.17148/IJARCCE.2026.153146
Abstract: Influencer marketing has become a major factor in consumerdecision-making due to the quick growth of social media, especially among Millennials and Generation Z. However, little is known about how influencer marketing affects purchase intention and brand trust, particularly in the Indian context. This study compares responses from Millennials and Gen Z and looks at how the three main components of influencer marketing relatability, authenticity, and credibility affect brand trust and purchase intention.A structured questionnaire was used to gather data from 312 active social media users in India as part of a quantitative research design. Descriptive statistics, reliability analysis, correlation, regression, mediation analysis, and independent samples t-tests were all used in the analysis of the data using SPSS. The results show that while influencer marketing has a relatively moderate effect on brand trust, it has a strong and significant impact on purchase intention. Although brand trust has a positive impact on purchase intention as well, its mediating effect isnot statistically significant, indicating that consumer decisions are directly influenced by influencer marketing. Furthermore, there was no discernible difference in the purchasing habits of Gen Z and Millennials.By highlighting the direct persuasive power of influencer marketing and providing useful insights for brands looking to engage digitally active consumers, this study adds to the growing body of literature on digital marketing.
Keywords: Millennials, Gen Z, Influencer Marketing, Purchase Intention, Authenticity, Credibility, and Relatability
Keywords: Millennials, Gen Z, Influencer Marketing, Purchase Intention, Authenticity, Credibility, and Relatability
Abstract
AutoMomo–Rapid Forming Technology
Ms.G.Madhumathi, Arini.R, Jayahari.S, Nilavarashi.T
DOI: 10.17148/IJARCCE.2026.153147
Abstract: Automation in the food processing industry has become increasingly important to improve productivity, maintain hygiene, and reduce labor dependency. Traditional momo preparation is performed manually, which is time- consuming and requires skilled labor to maintain uniform size and shape. This research focuses on the design and development of an automated momo making machine that integrates multiple motor-driven mechanisms to perform dough feeding, stuffing placement, molding, and product transfer operations.The developed system consists of four motor-controlled mechanisms and a conveyor belt system that performs sequential operations to produce uniformly shaped momos. Experimental trials were conducted at different operating speeds to evaluate the machine’smolding efficiency, product damage rate, and production capacity. Results show that the developed machine significantly improves production efficiency while maintaining consistent product quality. The optimum operating speed was determined to be 250 RPM, achieving a molding efficiency of 92% with minimal product damage. The proposed system is suitable for small-scale food industries and commercial kitchens where large-scale momo production is required.
Keywords: Food Processing Automation, Momo Making Machine, Dough Molding Machine, Conveyor System, Mechanical Food Processing.
Keywords: Food Processing Automation, Momo Making Machine, Dough Molding Machine, Conveyor System, Mechanical Food Processing.
Abstract
College Enquiry Chatbot UsingMachine Learning:An Intelligent Conversational System for Academic Information Retrieval
Sonu Yadav, Vivek Chavan, Sufyan Hawaldar, Samruddhi Chinchavale, Prof. Ashwini Chavan
DOI: 10.17148/IJARCCE.2026.153148
Abstract: TheCollege Enquiry Chatbotis an intelligent conversational system designed to automate the handling of student and visitor enquiries related to college admissions, courses, fees, faculty, facilities, and academic schedules. The system leveragesMachine Learning (ML), Natural Language Processing (NLP), and deep learning-based intent classificationto understand and respond to user queries in real time. Built using Python, NLTK, TensorFlow/Keras, and a Flask-based web interface, the chatbot delivers accurate, context-aware responses without human intervention. Unlike traditional FAQ pages or static information portals, this system understands the intent behind user queries, handles variations in phrasing, and provides personalized responses. The system is evaluated against benchmark metrics including accuracy, F1 score, and response latency. Experimental results demonstrate a classification accuracy of over 92% on a domain-specific college enquiry dataset. Future work includes multilingual support, voice integration, and CRM system connectivity.
Keywords: Artificial Intelligence, Chatbot, Deep Learning, Intent Classification, Machine Learning, Natural Language Processing, Neural Network.
Keywords: Artificial Intelligence, Chatbot, Deep Learning, Intent Classification, Machine Learning, Natural Language Processing, Neural Network.
Abstract
A Multimodal AI Framework for Real-Time Audience Engagement Detection in Virtual Communication
A V Tejaswi, M Sumana Sree, S Sahithi, Dr.C.Swapna
DOI: 10.17148/IJARCCE.2026.153149
Abstract: The way we globally collaborate has been significantly impacted by virtual communication tools. However, these tools are not effective in conveying nonverbal cues of engagement, which are important in the process of effective human interaction. Real-time audience attention evaluation is a problem that faces the presenter in virtual meetings and classrooms. This study discusses the AI-based methodology that can be used in detecting audience engagement via multimodal emotion recognition.The interactive Speaker Dashboard also displays individual participant engagement scores, which are determined by a Node.js backend server running a Video Analyzer, an Audio Analyzer, and an Engagement Engine. This proposed framework is applicable in various scenarios, such asremote training sessions, corporate meetings, virtual classroom scenarios, and customer support interactions. This framework also indicates better detection accuracy.This sensitivity to real participant behaviour can be seen in the experimental results, which demonstrate that the level of engagement aligns with the attentiveness pattern of live sessions.
Keywords: Engagement detection; Multimodal emotion recognition; Facial expression recognition; Speech emotion analysis; Affective computing; WebRTC; Virtual collaboration; Deep learning; Attention tracking.
Keywords: Engagement detection; Multimodal emotion recognition; Facial expression recognition; Speech emotion analysis; Affective computing; WebRTC; Virtual collaboration; Deep learning; Attention tracking.
Abstract
CROP WILTING ANGLE MEASUREMENT USING OPENCV SONIA MARIA D’SOUZA1, S VINAY KUMAR REDDY2, S SAMUEL SUNDAR3, SIDDAMREDDY MOHITH REDDY4, RAVVA SAI CHARAN5
Assoc. Professor, Department of AI & ML, New Horizon CollegeofEngineering, Bangalore, India
DOI: 10.17148/IJARCCE.2026.153150
Abstract: Water stress is one of the largest causes of crop yield loss in rainfed farming systems across dryland in rural areas. The early detection of water stress before the visible of discoloration or leaf death occurs, and it allows farmers to respond with targeted irrigation and can prevent yield losses of 30-60%. Current methods for detecting water stress rely on expensive soil moisture sensors, laboratory-based leaf water potential measurements or satellite imagery, all of these are hard to reach the smallholder farmers. This research proposes a solution for measuring the sloping angle of crop leaves and stems directly from a photograph taken with any basic smartphone camera. We use OpenCV based image processing to extract the structure of the plant's main stem or primary leaf, this compute the angle between the erect reference axis and the actual observed axis, and classify the resulting angle into three water stress categories called No stress (0–15 degrees), Mild stress (16–35 degrees) and severe stress (above 35 degrees) using a k-Nearest Neighbours classifier trained on annotated field images. Experiments can be conducted on major crops like (tomato, paddy, maize, sunflower, cotton, chilli etc.) The proposed method requires no specialist equipment, only a smartphone camera and a python script making it the most accessible water stress detection tool which is proposed for Indian smallholder farmers.
Keywords: Wilting Angle, Water Stress, OpenCV, rainfed farming, image processing, leaf water potential.
Keywords: Wilting Angle, Water Stress, OpenCV, rainfed farming, image processing, leaf water potential.
Abstract
IMPACT OF ORGANIZATIONAL POLITICS ON EMPLOYEE PERFORMANCE IN IT SECTOR IN PUNE
Abhishek Kolhe, Sayali Waghmare, Mrunali Shinde, Nandini Singharam, Pranjali Kirtane
DOI: 10.17148/IJARCCE.2026.153151
Abstract: Organizational politics has become a significant aspect of modern workplaces, especially in highly competitive sectors such as the Information Technology (IT) industry. This study examines the causes and effects of organizational politics and its impact on employee performance in IT companies located in Pune. Organizational politics arises due to factors such as favoritism, power struggles, lack of transparency in decision-making, and intense competition among employees for promotions and recognition.
These political behaviors influence employee perceptions and create an environment that may hinder productivity and collaboration. The effects of organizational politics are reflected in increased stress levels, reduced job satisfaction, lack of trust in management, and decreased employee performance. Employees who perceive high levels of politics tend to show lower engagement, reduced efficiency, and higher turnover intentions.
This study highlights the importance of implementing transparent HR practices, fair performance evaluation systems, and ethical leadership to minimize the negative impact of workplace politics.
Keywords: Organizational Politics, Employee Performance, IT Sector, Job Satisfaction, Workplace Behavior, Productivity
These political behaviors influence employee perceptions and create an environment that may hinder productivity and collaboration. The effects of organizational politics are reflected in increased stress levels, reduced job satisfaction, lack of trust in management, and decreased employee performance. Employees who perceive high levels of politics tend to show lower engagement, reduced efficiency, and higher turnover intentions.
This study highlights the importance of implementing transparent HR practices, fair performance evaluation systems, and ethical leadership to minimize the negative impact of workplace politics.
Keywords: Organizational Politics, Employee Performance, IT Sector, Job Satisfaction, Workplace Behavior, Productivity
Abstract
A Comparative Study of Cardiovascular Fitness between Football Players and Non-Football Players”
Vidya Bhushan Sharma
DOI: 10.17148/IJARCCE.2026.153152
Abstract: The present study was conducted to compare cardiovascular-related fitness between football players and non-football players of Swami Ramanand Teerth Marathwada University, Nanded. A total of 40 male subjects were selected and divided into two groups: football players (n = 20) and non-football players (n = 20). Speed was assessed using the 12-meter run test, while cardiovascular fitness was interpreted through performance outcomes associated with repeated high-intensity effort.
The results revealed that football players demonstrated superior performance in the 12-meter run, with a higher mean score (M = 2167.50, SD = 124.78) compared to non-football players (M = 1995.78, SD = 119.67). The obtained t-ratio (33.56) indicated a statistically significant difference between the two groups. These findings suggest that regular football training leads to better speed and enhanced cardiovascular efficiency due to continuous involvement in sprinting, acceleration, and recovery-based activities.
In conclusion, the study confirms that football players possess significantly better speed and cardiovascular fitness than non-football players. The observed differences highlight the positive impact of systematic football training on physical fitness and emphasize the importance of regular sports participation for improving cardiovascular health and performance among university students.
Keywords: Football players, Non-football players, Speed, 12-meter run, Cardiovascular fitness
The results revealed that football players demonstrated superior performance in the 12-meter run, with a higher mean score (M = 2167.50, SD = 124.78) compared to non-football players (M = 1995.78, SD = 119.67). The obtained t-ratio (33.56) indicated a statistically significant difference between the two groups. These findings suggest that regular football training leads to better speed and enhanced cardiovascular efficiency due to continuous involvement in sprinting, acceleration, and recovery-based activities.
In conclusion, the study confirms that football players possess significantly better speed and cardiovascular fitness than non-football players. The observed differences highlight the positive impact of systematic football training on physical fitness and emphasize the importance of regular sports participation for improving cardiovascular health and performance among university students.
Keywords: Football players, Non-football players, Speed, 12-meter run, Cardiovascular fitness