Volume - 12 Issue - 3

Area of research: Computer Science And Engineering

This Paper Presents DRIVE, A Placement Drive Rehearsal Interview Virtual Environment Designed To Simulate Real Campus Recruitment Processes. The System Enables Students To Practice Complete Placement Workflows By Integrating Aptitude Tests, Technical Interviews, Coding Assessments, And HR Interview Preparation Into A Single Platform. The System Allows Users To Enter A Company Name, Upon Which It Analyzes Recruitment Patterns And Generates A Structured Hiring Process. A Competition-based Elimination Model Is Implemented Using Dynamic Cut-off Values Derived From Simulated Candidate Pools. AI-driven Techniques Are Used To Provide Performance Analysis And Feedback. Unlike Traditional Mock Interview Systems, DRIVE Focuses On Full Placement Realism By Incorporating Time Constraints, Elimination Stages, And Performance Pressure. The System Aims To Improve Placement Readiness, Confidence, And Performance Of Students.

Author: Santhosh Ram S | Thiravidan P M | Sooriya S S | Mrs.N.Nazmunisha

Area of research: CSE

DevOps And MLOps Have Become Essential Skills, But Students Often Struggle To Navigate Large Amounts Of Unstructured Learning Material Such As Documentation, Blogs, And Tutorials. This Project Proposes An Agentic RAG-based Study Assistant That Helps Learners Understand DevOps And MLOps Concepts Through Domain-grounded Question Answering And Personalized Study Support. The System First Builds A Knowledge Base By Collecting And Chunking Trusted DevOps/MLOps Resources, Then Converts Them Into Embeddings And Stores Them In A Vector Database Using A Retrieval-Augmented Generation (RAG) Pipeline. On Top Of This, Multiple AI Agents Are Orchestrated: A Retrieval Agent That Selects Relevant Content, A Planner Agent That Creates Topic-wise Study Plans, And A Tutor Agent That Provides Explanations, Summaries, And Practice Questions. By Constraining Responses To Retrieved, Curated Material, The Assistant Aims To Reduce Hallucinations And Improve Factual Accuracy For Educational Us.

Author: K Mutheeswari | R.B.Vishnu | M.K. Vishwaraj | M.Srisivaraman

Area of research: Narcotic Drugs And Psychotropic Substances

Drug Abuse Continues To Be A Major Social And Legal Concern In Tamil Nadu Despite Government Claims Of Effective Drug Control. This Study Critically Examines The Gap Between The State’s Drug-free Narrative And The Actual Ground Reality. It Focuses On The Effectiveness Of Enforcement Policies, Public Awareness Programmes, And Rehabilitation Measures. The Research Also Considers Recent Drug-related Violence Involving Minors, Highlighting The Social Impact Of Substance Abuse. Primary Data Were Collected Through A Structured Questionnaire, While Secondary Data Were Obtained From Journals, Government Reports And Legal Provisions. The Findings Indicate That Strict Laws Alone Are Insufficient To Eliminate Drug Circulation. The Study Emphasizes The Need For A Balanced Approach Combining Enforcement, Prevention And Rehabilitation To Address Drug Abuse In Tamil Nadu Effectively.

Author: Yogadharshini.S | Dr. S. Thirumal

Area of research: Electronics And Communication Engineering

The Rapid Growth Of Surveillance Cameras In Public Places Has Led To The Generation Of Large Volumes Of Video Data, Making Manual Monitoring Difficult And Inefficient. Detecting Violent Activities Such As Fights, Weapon Usage, And Suspicious Behaviour In Real Time Is Essential For Maintaining Public Safety. This Paper Proposes A Deep Learning-based System For Crime Scene Analysis Using Multimodal Video Summarization And YOLO-based Object Detection. The System Processes Surveillance Videos By Extracting Frames And Applying Preprocessing Techniques, Followed By Object Detection Using The YOLO Algorithm. Convolutional Neural Networks (CNN) Are Used For Feature Extraction To Identify Objects Such As People And Weapons. The System Analyses Detected Objects To Identify Suspicious Activities And Generates Real-time Alerts With Relevant Information Such As Timestamps And Confidence Scores. This Approach Improves The Efficiency Of Surveillance Monitoring And Helps Security Personnel Respond Quickly To Potential Threats.

Author: Karwin Vikash Tr | Jean Benedict M | Iyappank | Subasree

Area of research: CSE

Air Pollution Is One Of The Most Critical Environmental Challenges Affecting Urban And Industrial Regions Worldwide. Rapid Industrialization, Vehicular Emissions, And Construction Activities Have Significantly Increased The Concentration Of Harmful Pollutants In The Atmosphere. This Paper Presents A Comprehensive Study On Air Pollution Control Systems, Including Their Design, Working Principles, And Effectiveness. Various Control Technologies Such As Electrostatic Precipitators, Cyclone Separators, Scrubbers, And Filtration Systems Are Analyzed. The Proposed System Integrates Multiple Pollution Control Techniques To Achieve Higher Efficiency In Removing Particulate Matter And Gaseous Pollutants. The Study Also Evaluates System Performance Based On Efficiency, Cost, And Environmental Impact. The Results Indicate That A Hybrid Air Pollution Control System Can Significantly Reduce Airborne Contaminants And Improve Air Quality, Contributing To Sustainable Development..

Author: Atharva Sandip Shinde | Gaurav Ajay Naidu | Akshay Tukaram Jadhav | Pratik SudarshanIngale | Mr. Shendge Sir

Area of research: Electronics And Communication Engineering

Rapid Urbanization Has Led To A Surge In Vehicle Ownership, Resulting In Significant Parking Management Crises Characterized By Heavy Congestion And Time Loss. This Paper Proposes An Automated Smart Parking System That Leverages CCTV Technology And Infrared (IR) Sensors To Optimize Facility Oversight. By Employing Advanced Image Processing Techniques, The System Monitors Vehicle Entry And Exit Points While Identifying Real-time Slot Occupancy. Data Is Managed Via A Centralized Server And Communicated To Motorists Through A Web-based Interface. Furthermore, The System Enhances User Experience By Providing Automated SMS Alerts Regarding Slot Assignments, Duration Of Stay, And Billing. The Results Demonstrate That This Integrated Approach Minimizes The Need For Manual Intervention, Increases Operational Efficiency, And Offers A Scalable Solution For Modern Smart City Infrastructures.

Author: Dr.P. Elayaraja | M. Pradeep | B. Prathap | K. Saravanan

Area of research: Psychology

This Study Investigates How Attachment Insecurity Predicts Identity Problems In Young People. Identity Building Is An Important Developmental Job In Emerging Adulthood, Yet Many People Experience Instability In Their Self-concept And Emotional Functioning. Attachment Instability, Which Originates In Early Caregiving Experiences, Has A Substantial Impact On Interpersonal Relationships And Self-perception. The Study Included 300 Young Individuals Aged 18-25 Who Took Standardized Measurements. The Findings Revealed A Substantial Positive Link Between Attachment Insecurity And Identity Disturbance. The Findings Underline The Significance Of Early Relational Experiences In Identity Formation, As Well As The Necessity For Treatments That Promote Emotional Control And Secure Connections.

Author: Maragatha Priyanka P | Mr. Manoj R | Ms. Jayashree

Area of research: Psychology

The Study Looks At The Connection Between Emotional Dysregulation And The Desire To Injure Oneself In Young Adults Between The Ages Of 18 And 25. Descriptive Statistics And Pearson's Correlation Were Used To Examine The Data Using A Sample Of 350 Individuals And Standardized Instruments (DERS And NSSI-AT).The Results Showed A Weak And Non-significant Correlation Between Low Self-harm Impulses And Moderate Emotional Dysregulation. This Suggests The Impact Of Additional Psychological And Social Factors, As Emotional Dysregulation By Itself Does Not Significantly Predict Self-harm Urges In Non-clinical Groups.:

Author: Keerthana R | Mr. Manoj R | Ms. Shalini R

Area of research: Artificial Intelligence

Large Language Models (LLMs) Often Generate Plausible Yet Incorrect Information, Known As Hallucinations. This Paper Proposes A Real-time Hallucination Detection System That Evaluates The Reliability Of LLM Outputs. The System Combines Evidence Retrieval From Trusted Sources, Semantic Similarity Using Sentence Embeddings, And Self-consistency Checks Across Multiple Responses. A Unified Decision Module Classifies Outputs As Factual Or Hallucinated. Implemented As A Streamlit Web Application, The System Provides An Intuitive Interface For Evaluating Responses. This Approach Enhances Transparency, Reliability, And Trust In AI-generated Content For Research And Professional Use.

Author: Keerthi .K | Jayashree.S | Dharani.R | Archana.P | Mrs.J. jenila

Area of research: Computer Science And Engineering

This Project Showcases The Design And Deployment Of A Multi-tier, Enterprise-grade Cloud Architecture On AWS That Emphasizes High Availability, Scalability, And Security. The Infrastructure Integrates A Wide Range Of Managed AWS Services — Including VPC, EC2, ELB, RDS, EFS, S3, CloudFront, Lambda, CloudWatch, And SNS — To Deliver A Fully Automated And Resilient Web Application Environment. Leveraging Multi-AZ Deployment And Auto Scaling, The Architecture Ensures Seamless Performance And Fault Tolerance During Variable Workloads. Security Is Reinforced Through Network Segmentation Using Public And Private Subnets, Fine-grained IAM Policies, And VPC Peering For Secure Inter-network Communication. CloudFront CDN Accelerates Content Delivery Globally, While CloudWatch And Lambda Enable Intelligent Monitoring And Automated Incident Response. Designed In Alignment With The AWS Well-Architected Framework, The Solution Demonstrates Excellence In Operational Efficiency, Performance Optimization, And Cost-effective Cloud Management, Making It A Scalable Foundation For Modern Enterprise Applications.

Author: Saranya P | Hariprasath M

Area of research: Computer Science And Engineering

Agriculture Plays A Huge Role In The Economies Of Developing Nations Like India, Where It Keeps More Than Half The Workforce Employed. Yet, Despite Its Importance, Farmers Still Struggle With How They Use Chemical Fertilizers. Too Much Fertilizer Degrades The Soil, Pollutes Lakes And Rivers, And Costs Farmers A Fortune. Too Little? Crops Barely Grow, And Yields Drop.This Paper Introduces An IoT-powered, Smart Crop Fertilizer Recommendation System Using Machine Learning, Designed To Tackle The Problem Head-on. At The Heart Of The Setup Is An ESP8266 NodeMCU V3 Microcontroller, Which Connects To A DHT11 Sensor For Temperature And Humidity, Plus A Capacitive Sensor To Check Soil Moisture. Together, These Monitor Key Environmental And Soil Conditions Around The Clock — Temperature, Humidity, And How Much Moisture The Soil Holds. Alongside These Readings, The System Collects Soil Nutrient Levels (nitrogen, Phosphorus, Potassium) And Sends Everything Wirelessly To A Cloud Server.On The Server, A Trained XGBoost Model Analyzes The Data And Delivers A Clear, Targeted Recommendation: The Best Fertilizer, The Right Blend, And The Exact Dosage. Farmers Can Check These Recommendations In Real Time Through A Web Dashboard Built With React.js And Tailwind CSS. It’s Simple And Friendly, So Anyone Can Navigate It Without Fuss.Tests Show The System Reaches 93.6% Accuracy In Its Classifications And Responds In Under A Second — About 900 Milliseconds — From Data Collection To Recommendation. This Proves It Has Real Potential For Smarter, More Precise Farming At Scale.

Author: Ganeshen P | Rasiga Priya M | Priyadharshika M | Sathya Sri P V | Srivarshini R

Area of research: Computer Science And Engineering

Gas Leaks Are A Real Threat—one That Puts Lives, Property, And Industries At Constant Risk. Fires, Explosions, Toxic Exposure, Even Simple Kitchen Accidents Trace Back To Leaks That Go Unnoticed For Just A Few Minutes. The Old Ways Of Handling This—standalone Detectors, Periodic Checks, Or Wired Alarms—just Don’t Cut It Anymore. They Are Slow, Limited To One Spot, And Often Miss The Bigger Picture, Especially In Sprawling Industrial Plants Or Crowded Urban Homes. This Work Introduces A Better Solution: An IoT-Enabled Unified Gas Safety Management System. Here How It Works. Using MQ-series Gas Sensors Linked To An Arduino Nano, The Platform Tracks Hazardous Gases, Monitors Temperature Changes, And Detects Open Flames—all At Once. Everything Happens In Real-time. When Sensors Spot A Problem—say, Rising Gas Levels Or Unexpected Heat—the System Doesn’t Just Beep. It Powers Alarms, Shuts The Gas Valve, Starts The Exhaust Fan, And Instantly Updates The Status On A Local Screen. Meanwhile, All Sensor Readings Get Sent Wirelessly, So Operators (no Matter Where They Are) Know What’s Happening And Can Respond Instantly. Testing This Setup Produced Real Results: It Reliably Detected Gas, Kept Errors Under 12 Ppm, And Triggered Alerts In Just 1.3 Seconds. There Were No False Positives, And The System Stayed Up And Running 100% Of The Time Throughout A 72-hour Test. It’s Affordable, Scalable, And Works Just As Well In A Factory As It Does In A Home Or Store. This Design Gives A Unified Approach To Gas Safety—way Beyond Fire Drills And Manual Checks.

Author: Balaji Saravanan U K | Rishwanth M | Sakthivel R | Santhosh S V

Area of research: Computer Science And Engineering

The Rapid Growth Of Online Banking Has Significantly Increased Fraudulent Activities Such As Unauthorized Transactions, Phishing Attacks, Identity Theft, And Account Takeovers. Traditional Rule-based Systems Fail To Detect Evolving Fraud Patterns. This Paper Proposes A Machine Learning-based Fraud Detection System That Analyzes Transaction Behavior And Identifies Anomalies In Real Time. The System Employs Logistic Regression, Decision Tree, Random Forest, And K-Nearest Neighbors Algorithms, Achieving Up To 95% Detection Accuracy. Results Demonstrate Significant Improvement In Accuracy, Reduction In False Positives, And Enhanced Banking Security Compared To Conventional Approaches.

Author: Nagarajan V | Nandakumaran M | Naveen Kumar C | Mrs. S. Ramalakshmi

Area of research: Artificial Intelligence And Data Science

Blockchain Technology Has Revolutionized Distributed Systems Through Its Decentralized, Immutable, And Transparent Architecture. However, The Increasing Adoption Of Blockchain Networks Has Attracted Malicious Actors Exploiting Vulnerabilities For Fraud, Money Laundering, And Other Illicit Activities. This Paper Presents A Comprehensive Machine Learning-based Framework For Detecting Anomalies Across Multiple Layers Of Blockchain Architecture. We Propose A Multi-layered Detection System That Integrates Supervised, Unsupervised, And Deep Learning Techniques To Identify Suspicious Patterns In Transaction Flows, Smart Contract Execution, And Network Behavior. Our Evaluation On Bitcoin And Ethereum Datasets Demonstrates 94.7% Detection Accuracy With A False Positive Rate Of 2.3%. The Proposed System Addresses Key Challenges, Including Limited Labeled Data, Real-time Processing Requirements, And Privacy Preservation Through Federated Learning Integration.

Author: Suvitha S | Kiruthika S | Divish S | Vignesh R | Sivashankar R

Area of research: AI And Deep Learning In Medical

Eye Diseases Such As Diabetic Retinopathy (DR), Glaucoma, And Age-related Macular Degeneration (AMD) Remain Major Causes Of Vision Loss Globally. Early Screening Is Crucial But Limited By The Availability Of Trained Ophthalmologists. This Paper Proposes A Deep Learning–based Automated Screening System For Multiclass Eye Disease Detection Using Retinal Fundus Images. The Model Integrates Preprocessing Enhancement, Feature Extraction Using CNN Architectures, And Classification Using Transfer Learning. Experiments Conducted Publicly Available Datasets Demonstrate High Diagnostic Accuracy, Outperforming Conventional Models. This Work Highlights The Potential Of Deep Learning For Scalable And Cost-effective Ophthalmic Disease Screening.

Author: Mrs.I.Suganya | Mrs. R. K. Nithya | Ms.K.Kavipriya | Ms.A.Sindhupriya | Ms.S.T.Swathi

Area of research: Artificial Intelligence In Healthcare And Pharmacy

The Rapid Advancement Of Digital Health Technologies Has Created New Opportunities To Transform Traditional Pharmacy And Patient Medication Management Systems Into Intelligent, Interconnected Platforms. The AI-Enhanced Pharmacy Medi-Track System Is A Comprehensive Healthcare Management Solution Engineered To Automate And Optimize Core Processes, Including Online Physician Appointment Scheduling, Smart Medicine Tracking, AI-driven Drug Recommendations, Electronic Prescription Management, And Real-time Medication Intake Alerts. Built Using Python, Flask, And A React-based Frontend, The System Integrates Machine Learning Models Trained On Clinical Datasets To Generate Personalized Drug Suggestions Based On Patient History, Diagnosed Conditions, And Allergy Profiles. A Dedicated Notification Engine Leverages Push Alerts And SMS Gateways To Ensure Patients Adhere To Prescribed Regimens. The Backend Is Supported By A PostgreSQL Relational Database, Enabling Secure And Scalable Storage Of Patient Records, Transactional Pharmacy Data, And Longitudinal Health Histories. Furthermore, A Built-in Analytics Dashboard Provides Healthcare Providers With Actionable Insights Derived From Aggregated Patient Data. This Research Presents The Architecture, Design Rationale, Implementation Strategy, And Performance Evaluation Of The Proposed System, Demonstrating Its Potential To Reduce Medication Errors, Enhance Patient Compliance, And Streamline Pharmaceutical Workflows Across Diverse Healthcare Environments.

Author: Mrs. S. R. Saranya | Sivashakthi K | Ragul G | Mani G

Area of research: Civil Engineering

Beam Column Joints Are Critical Zones In Reinforced Concrete (RC) Framed Structures That Significantly Influence Seismic Performance. Under Earthquake Induced Cyclic Loading, Joints Are Subjected To Complex Stress States Including Shear, Bending, And Bond Stresses, Leading To Stiffness Degradation, Energy Dissipation, And Eventual Failure. This Experimental Study Investigates The Behavior Of RC Beam Column Joints Under Reversed Cyclic Loading. A Total Of Three Specimens With Varying Reinforcement Detailing Were Tested Under Controlled Laboratory Conditions. Key Parameters Such As Load Displacement Response, Stiffness Degradation, Energy Dissipation, Ductility, Crack Patterns, And Failure Modes Were Evaluated. Results Show That Joint Detailing Substantially Affects Seismic Performance, And Strengthened Joints With Adequate Confinement Exhibit Higher Ductility And Energy Absorption Capacity. The Findings Provide Insights For Improved Seismic Design And Retrofitting Strategies.

Author: Aditya Dawale | Mayur Lanjulkar | Prasad Ghongate | Shubham Wanare | Rushikesh Khodke | Prof. M. S. Sindhikar

Area of research: Civil Engineering

Urban Infrastructure Projects Play A Critical Role In Shaping Economic Growth, Social Welfare, And Environmental Health In Rapidly Urbanizing Regions. Sustainable Evaluation Of Such Projects Requires Systematic Assessment Across Multidimensional Indicators That Balance Social, Environmental, And Economic Impacts. This Paper Presents A Comprehensive Review And Framework For Evaluating The Sustainability Of Urban Infrastructure, Integrating Methodologies Such As Multi Criteria Analysis, Indicator-based Scoring, And Expert Weighting Systems. It Synthesizes The Latest Approaches In Sustainability Assessment And Proposes Practical Recommendations For Planning, Monitoring, And Policy Decision Making, Facilitating More Resilient And Equitable Urban Development.

Author: Abhishek Tayade | Prashant Baviskar | Sayyam Jain | Nimbaji Tayde | Gaurav Ghadekar | Prof. P. K. Patil

Area of research: Artificial Intelligence For Space Application

The Rapid Increase In Satellite Launches And Orbital Debris Has Raised Critical Challenges In Ensuring Collision-free Operations And Efficient Launch Scheduling. ISO-AI Lite Is A Lightweight, Explainable Artificial Intelligence Pipeline Designed To Assist In Launch-window Validation, Conjunction Risk Assessment, And Minimal Avoidance Manoeuvre Planning. The System Takes Two-Line Element (TLE) Data Of A Satellite And A Potential Conjunction Object As Input, Propagates Their Orbits Using The SGP4 Model, And Estimates The Probability Of Collision (PoC) Through Both Analytical And Monte Carlo Methods. It Integrates A Simple Weather Constraint Checker To Validate Launch Feasibility Based On Basic Environmental Parameters. When A High-risk Conjunction Is Detected, The Model Suggests An Optimal, Low-cost Avoidance Manoeuvre, Preferably In The Along-track Direction, Ensuring Both Safety And Fuel Efficiency. Additionally, The Framework Includes An Optional Single-sensor Re-observation Module That Refines Orbital Uncertainty Through A Kalman-based Update Before Executing The Manoeuvre. The Overall Objective Of ISO-AI Lite Is To Demonstrate A Compact, Interpretable, And Cost-effective Decision-support Tool For Mission Operators And Students, Enabling Improved Situational Awareness, Risk Mitigation,

Author: Dr.P.Thangavel | Niveditha A | Adithyan Sabu | Mohammed Nisar | Nirmal Kumar

Area of research: Civil Engineering

The Project “Production Of Manure Using Waste” Focuses On Converting Domestic Organic Waste Into Nutrient-rich Manure Through A Simple And Eco-friendly Process. The Primary Aim Is To Reduce Household Waste Such As Fruit And Vegetable Peels, Spoiled Dairy Products, And Dry Leaves, While Simultaneously Producing An Effective Organic Fertilizer. The Methodology Involves Collecting Biodegradable Waste And Decomposing It In A Layered Container System Over A Period Of 35–40 Days. The Prepared Manure Was Tested On Selected Plants, Including Tomato And Flower Saplings, And Its Performance Was Compared With Chemical Fertilizers And Normal Watering Conditions Over A Span Of 21 Days. The Results Indicated That While Chemical Fertilizers Showed Slightly Faster Growth In Some Cases, The Organic Manure Provided Sustainable And Healthy Plant Development. Additionally, The Use Of Manure Improves Soil Fertility, Moisture Retention, And Microbial Activity. This Project Demonstrates An Efficient Way To Manage Domestic Waste And Promote Sustainable Agricultural Practices By Reducing Reliance On Chemical Fertilizers And Encouraging Environmentally Friendly Alternatives.

Author: Ms Farah Anjum Shaikh | Swapnil Bandgar | Meet Ukani | Shivam Kale | Aditya Boyale

Area of research: Civil Engineering

The Study On Seasonal Variations Of Water Parameters In Hadapsar Analyzes Changes In PH, Turbidity, Total Hardness, And Dissolved Solids Across Summer, Monsoon, And Winter Seasons. Results Indicate Higher Turbidity During Monsoon Due To Runoff, While PH And Hardness Show Noticeable Fluctuations In Winter And Summer. The Findings Help Assess Water Quality Status And Its Suitability For Drinking And Civil Engineering Applications In The Region.

Author: Mr.A.Masal | Suresh Kakde | Vaibhav Kamble | Yash Ovhal | Ajay Kakde

Area of research: Artificial Intelligence

In Recent Years, Artificial Intelligence Systems Have Evolved Significantly, Enabling Interaction Through Multiple Models Such As Natural Language Processing (NLP), Computer Vision, And Speech Processing. However, Most Existing Systems Operate In Isolation, Leading To Inefficiencies, Redundancy, And Lack Of Precision In Responses. This Paper Proposes A Precision-Optimized Multi-Model Interaction And Response Synthesis Framework, Which Integrates Multiple AI Models Into A Unified System To Enhance Accuracy, Contextual Understanding, And Response Quality. The Proposed Framework Dynamically Selects And Coordinates Multiple Models Based On User Input, Context, And Task Requirements. It Employs An Intelligent Orchestration Mechanism That Analyzes Input Data, Routes It To Appropriate Models, And Synthesizes Outputs Into A Coherent And Optimized Response. The System Also Incorporates Feedback Mechanisms To Continuously Improve Performance. Experimental Results Indicate That The Proposed System Significantly Improves Response Precision, Reduces Latency, And Enhances User Experience Compared To Traditional Single-model Systems. This Framework Can Be Applied In Domains Such As Healthcare, Education, Virtual Assistants, And Smart Automation Systems.

Author: Deepak S | Mohamed Thameem S | Mithesha S | Hemalatha C | Mrs. V. Gomathi

Area of research: Artificial Intelligence And Data Science

Yoga Has Emerged As A Vital Practice For Enhancing Physical Fitness, Mental Well-being, And Overall Health. However, Practicing Yoga Without Proper Guidance Can Lead To Incorrect Postures, Reducing Effectiveness And Increasing Injury Risk. This Paper Presents Yoga AI, An Intelligent System Leveraging Artificial Intelligence And Computer Vision For Real-time Posture Correction And Asana Guidance. The System Employs MediaPipe Pose Estimation Framework To Capture Body Keypoints From Video Input, Calculates Joint Angles To Assess Posture Accuracy, And Provides Immediate Visual And Auditory Feedback And Underatnding Of Integrated With Deep Learning Models Including Convolutional Neural Networks (CNNs), The System Achieves 94.3% Accuracy In Pose Classification Across 15 Common Yoga Asanas. Experimental Results Demonstrate The System's Effectiveness In Detecting Postural Deviations And Guiding Users Toward Correct Alignment. The Proposed Solution Addresses The Growing Need For Accessible, Personalized Yoga Instruction In Digital Wellness Platforms, Rehabilitation Centers, And Home-based Practice Environments.

Author: Mr.Anand M | Mouleeswar SV | Visvak Sena D | Pranesh T

Area of research: Civil Engineering

A Water Tank Model To Study The Behavior Of Waves And Their Impact On Coastal Areas. Artificial Waves Are Generated In The Tank To Simulate Sea Waves And Flood And A Sloping Bed And Coastal Defence Structures (such As Seawalls, Steps, Rock Amorus, Re-curve Wall) Are Modeled. The Experiment Helps In Understanding Wave Propagation, And The Effectiveness Of Coastal Protection Measures. Such Physical Modeling Is Useful For Designing Safe And Efficient Coastal Defence Structures, Reducing Flood Risk, And Protecting Coastal Regions From And Natural Disasters

Author: Mr. Nishchay More | Dhanashree Kadam | Ruchi Chavan | Prachi Vishwekar | Krish Gholave | Sushant Chaudhari

Area of research: Civil Engineering

In Recent Years, The Demand For Sustainable And Energy-efficient Building Materials Has Increased Due To Rising Temperatures And Environmental Concerns. This Study Focuses On The Development Of Natural Cooling Bricks Using Aloe Vera Mixed With Cement. Aloe Vera Is Known For Its High Water Retention And Cooling Properties. The Prepared Bricks Are Tested For Temperature Reduction, Water Absorption, And Compressive Strength. Results Show That Aloe Vera Bricks Maintain Lower Temperature Than Normal Bricks.

Author: Ms.Priti.E.Sangode | Randive Abhinav Pravin | JagtapTanish Vilas | Jadhav Aniket Ganesh | Kshirsagar Swayam Nitin

Area of research: ECE

Driving In Hill Stations And Mountainous Regions Is Highly Challenging Due To Frequent Fog, Mist, And Sharp Curves, Which Significantly Reduce Road Visibility And Increase The Risk Of Accidents. To Address This Issue, This Project Proposes A Smart Driver Assistance System That Enhances Vehicle Safety Under Low Visibility Conditions. The System Integrates Distance-based Vehicle Detection Using Ultrasonic Sensors With Real-time Image Dehazing Techniques For Improved Road Monitoring. Ultrasonic Sensors Are Mounted At The Front And Rear Of The Vehicle To Continuously Measure The Distance Of Nearby Vehicles. The Measured Distances Are Displayed In Meters On A Visual Display Unit, And An Audible Warning Is Generated Using A Buzzer When Any Vehicle Approaches Within A Critical Range Of 10–12 Meters. This Provides Timely Alerts To The Driver And Helps Prevent Collisions. In Parallel, A Dash Camera Captures Real-time Video Of The Road Ahead, Which Is Processed Using An Image Dehazing Algorithm To Reduce The Effects Of Fog And Mist, Thereby Improving Visibility. The Proposed System Offers A Cost-effective And Reliable Solution For Enhancing Driving Safety In Fog-prone Hill Areas. By Combining Sensor-based Distance Monitoring With Vision-based Dehazing, The System Assists Drivers In Making Safer Driving Decisions Under Adverse Weather Conditions.

Author: Mr Prabakaran J | Arut Selvan S | Caroll Sherrwin C | Hari Vignesh I | Arithas M

Area of research: Artificial Intelligence And Web Development

N Effective Collaboration Process In Synchronous Software Engineering Education Has Always Posed A Great Challenge, Especially When Working With Student Teams. Current Collaborative Tools, Including Version Control Tools, Have Failed To Provide The Required Real-time Pedagogical Oversight, Which Enables The Monitoring Of Student Contributions Towards The Software Development Process. In This Paper, We Propose A Novel Intelligent Multi-agent System Called Collabify AI, Which Aims At Improving Collaborative Workspaces With Real-time Monitoring And Verification. We Have Used A Sensor-based Approach, Coupled With Large Language Models, To Monitor The Software Development Process, Thus Maintaining A Dynamic Project Meta-model. In The Proposed Methodology, We Have Used The Concept Of AI Based "Automated Verification Engine" (AVE), Which Continuously Verifies The Real-time Construction Of The Software System With Respect To The Predefined System Design. The Experimental Results Show The Effectiveness Of The Proposed System, As It Increases The Project Awareness Of The Student Team By 35% While Reducing Architectural Inconsistencies. Additionally, The Proposed System Allows The Student Team, Consisting Of Four Members, To Make A Fair Contribution Mapping With The Help Of The AI-based Dashboard. We Conclude That The Transition To Active And Agentic AI Intervention Is A Promising Solution To Tackle Complex Software Engineering Projects In Educational Contexts. It Is The Foundation For Future Autonomous Pair Programming.

Author: Jain Prasannakumar | Jassim Mohammed | JomGeo George | R Arjun | Ms. M. Sheeba

Area of research: CSE

This Project Presents PhishNet AI, A Proactive And Intelligent Mobile-based Security Framework Designed To Detect And Prevent Phishing Attacks Caused By Malicious URLs. With The Rapid Increase In Digital Communication Through SMS, Emails, And Social Media Platforms, Phishing Attacks Have Become A Major Cybersecurity Threat. The Proposed System Analyzes URL-based, HTML-based, And Derived Features Using Advanced Feature Engineering And Machine Learning Models To Classify URLs As Legitimate Or Malicious Before User Interaction. Unlike Traditional Systems That Rely On Reactive Detection, PhishNet AI Performs Real-time Analysis And Blocks Harmful Links At An Early Stage. The System Integrates Multiple Security Mechanisms Such As Web Application Firewall (WAF), Intrusion Detection And Prevention Systems (IDS/IPS), Bot Detection, And Secure Communication Using TLS And OAuth/JWT Authentication. It Is Optimized For Mobile Devices, Ensuring Low Latency, Minimal Resource Consumption, And High Accuracy.

Author: A.Nandhini | G.Gokulkannan | M.Sabeena | J.Saranya | J.Shehara Banu

Area of research: Electronics And Communication Engineering

In Modern VLSI Design, Arithmetic Circuits Are Fundamental Components That Directly Influence The Performance, Power Consumption, And Silicon Area Of Digital Systems. The Carry Select Adder (CSLA) Is One Of The Most Widely Used High-speed Adder Architectures Due To Its Ability To Compute Partial Sums In Parallel For Both Possible Carry Input Conditions. However, The Conventional CSLA Employs Two Complete Ripple Carry Adder (RCA) Units Per Group, Resulting In Significant Area Overhead And Increased Power Dissipation. This Paper Presents A 32-bit Optimized CSLA Architecture That Replaces The Second RCA In Each Group With An Explicitly Hardcoded Gate-level Binary To Excess-1 Converter (BEC) Incorporating Common Boolean Logic (CBL) Optimization. The BEC Module Computes The Increment-by-one Operation Using Shared AND Terms, Eliminating Redundant Logic Operations And Reducing Switching Activity Across The Design. The Proposed Architecture Is Implemented In Verilog HDL And Synthesized On Xilinx Artix-7 FPGA (xc7a100tcsg324-1) Using Vivado Design Suite. Synthesis Results Confirm That The Proposed Design Achieves 14.3% Reduction In Area, 11.1% Reduction In Logic Power, And 6.9% Improvement In Area-Delay Product (ADP) Compared To The Conventional Dual-RCA CSLA, Making It Highly Suitable For Low-power IoT Devices, Embedded Systems, And Battery-operated Portable Electronics.

Author: Prasanth E | Seshadri S | Boopathi S | Yazhini K

Area of research: Commerce

The Rapid Growth Of Digital Technology Has Transformed The Way Individuals Participate In The Share Market. Social Media Platforms, Including YouTube, Twitter, And Telegram, Have Emerged As Important Sources Of Financial Information And Investment Guidance. This Study Examines The Impact Of Social Media On Investor Behaviour In The Stock Market. It Highlights How Real-time Information, Opinions, And Trends Shared On These Platforms Influence Investment Decisions. While Social Media Improves Awareness And Accessibility, It Also Introduces Risks Such As Misinformation, Herd Behaviour, And Emotional Decision-making. The Study Aims To Analyse Both The Positive And Negative Effects Of Social Media On Investors, Emphasising The Importance Of Informed Decision-making In A Digital Environment.

Author: Mrs S. Maria Melka

Area of research: Computer Engineering

This Project, Titled “A Safety Zone Vehicle Parking Management System Is A Digital Solution Designed To Streamline And Optimize Parking Operations In Urban Areas, Commercial Spaces, And Institutions. The System Helps In Automating Vehicle Entry, Exit, Slot Allocation, And Fee Calculation, Reducing Human Effort And Errors. The Primary Objective Of This Project Is To Develop An Efficient And User-friendly Parking Management System Using Modern Technologies. The System Will Allow Users To Check For Available Parking Spaces, Book Slots In Advance, And Make Cashless Payments. It Will Also Assist Administrators In Monitoring Parking Usage, Preventing Unauthorized Access, And Generating Reports For Better Decision-making. Key Features Of The System Include Real-time Slot Availability Tracking, Vehicle Registration, Automatic Ticket Generation, And A Secure Database For Storing Vehicle Details. Technologies Spaces In Advance. Such As LoT Sensors, Machine Learning, And Cloud-based Storage Can Be Integrated For A More Advanced Approach. This Project Aims To Enhance Convenience, Security, And Efficiency In Vehicle Parking, Reducing Congestion And Saving Time For Both Users And Parking Operators..

Author: Todkar Ashwini | Satpute Riya | Mane Mayuri | Bansode Tanuja | Mr. Sugre D.D

Area of research: Biosignal Processing

Parkinson's Disease (PD) Affects Gait Dynamics, And Hence, Objective Analysis Is Required For Accurate Diagnosis. In This Regard, This Manuscript Proposes A Deep Learning Framework For Modeling Parkinsonian Gait Using Vertical Ground Reaction Force (VGRF) Signals. A Hybrid CNN-LSTM Network Is Used To Effectively Capture Spatial And Temporal Features Of The Gait, And A 16-dimensional Latent Space Is Used To Effectively Capture Discriminative Gait Features. An Accuracy Of 99%, Along With High Precision And Recall, Is Achieved By The Network, And A High AUC Of 0.999 Indicates Effective Separability Of Classes. Furthermore, Using Principal Component Analysis On The Learned 16D Space, Distinct Clusters Of Healthy And Parkinsonian Gait Patterns Are Observed. It Is Thus Concluded That The Proposed Framework Is Effective For Accurate Detection Of PD Using Gait Analysis.

Author: Lumen Christy V | R Amith Raj Kumar

Area of research: Management Studies

Startup Culture Among Indian Youth Startup Culture Has Become A Significant Driving Force In The Economic And Social Transformation Of India. In Recent Years, Indian Youth Have Increasingly Shown Interest In Entrepreneurship As A Career Option Rather Than Traditional Employment. Factors Such As Technological Advancement, Digital Platforms, Supportive Government Initiatives Like Startup India, And Easier Access To Funding Have Encouraged Young Individuals To Launch Innovative Startups. The Startup Ecosystem Has Created Opportunities For Creativity, Risk-taking, And Self-employment Among The Younger Generation. This Study Focuses On Understanding The Growth Of Startup Culture Among Indian Youth, The Motivating Factors Behind Their Entrepreneurial Intentions, And The Challenges They Face While Establishing Startups. Key Influences Include Education, Exposure To Entrepreneurial Role Models, Availability Of Incubation Centers, And The Rapid Growth Of The Digital Economy. At The Same Time, Issues Such As Financial Risk, Lack Of Experience, Market Competition, And Regulatory Challenges Continue To Affect Young Entrepreneurs. The Findings Highlight That Startup Culture Not Only Promotes Innovation And Economic Growth But Also Contributes To Job Creation And Skill Development Among Youth. Encouraging Entrepreneurship Through Training Programs, Financial Support, And Policy Initiatives Can Further Strengthen The Startup Ecosystem In India And Empower Young People To Become Job Creators Rather Than Job Seekers.

Author: P. Sowmiya | Dr. R. Rathidevi

Area of research: Internet Of Things

Custodial Deaths In India, Often Stemming From Torture Or Neglect During Arrest And Detention, Undermine Police Accountability And The Constitutional Right To Life Under Article 21. With Annual Averages Of 92–150 Incidents Over Two Decades, These Tragedies Disproportionately Burden Marginalized Groups, Exacerbated By Oversight Gaps Like Incomplete Closed-circuit Television Coverage In The High-risk Arrest-to-lock-up Phase. This Study Proposes IoT-enabled Smart Handcuffs As A Proactive Solution For Humane Policing, Integrating Vital Sign Monitoring Via Photoplethysmography For Heart Rate And Accelerometers For Activity Detection, Alongside Global Positioning System Tracking And Tamper-proof Encryption Through ESP32 Firmware. The Device Activates Upon Cuffing, Streaming Real-time Data To Cloud Servers And Alerting Officers To Anomalies Such As Distress Or Collapse, Ensuring Tamper-proof Logs For Evidentiary Use. Prototyping And Simulations Demonstrate Seamless Integration With Existing Protocols, Projecting A 40–60% Drop In Fatalities. Ultimately, This Innovation Scales Affordably Across 16,000+ Stations, Bolstering Article 21 Protections And Aligning With Global Anti-torture Norms To Foster Transparent, Ethical Law Enforcement.

Author: Praveen kumar R | S Keerthika | Kalaiyarasan K

Area of research: Computer Science And Engineering

Healthcare Smart Support Is An Artificial Intelligence (AI)–based Healthcare Support Management System Developed Using The Salesforce Customer Relationship Management (CRM) Platform And A Python-based Natural Language Processing (NLP) Engine. Patients Submit Requests Through The Web-to-Case Feature In Salesforce CRM, Which Automatically Creates Case Records. Non-clinical Requests Such As Billing, Insurance, Fraud, And Service Issues Are Converted Into Feedback Records For Sentiment Evaluation, While Appointment Requests Require Selection Of Medical Specialization And Are Routed Automatically To The Appropriate Medical Queue.A Scheduled Apex Batch Process (Salesforce Server-side Automation Program) Securely Sends Feedback Text To A Python NLP Service Hosted On The Python Anywhere Cloud Platform. The System Uses The VADER (Valence Aware Dictionary And SEntiment Reasoner) Model To Analyze Sentiment Score, Sentiment Label (Positive, Negative, Neutral), Escalation Probability, And Healthcare Keywords. I Developed An AI-based Healthcare Support System That Analyzes Patient Data To Identify Risk Levels. The Results Are Returned To Salesforce CRM, Where Predictive Insights Are Generated And Alerts Are Sent Automatically To Improve Patient Care And Support Efficiency.

Author: Hemalatha K | Santhiya S | Monish Kumar S | Prof. Mrs. X. Anitha Sarafin | Prof.Dr.A Geethapriya

Area of research: ECE

We Propose A Novel Foggy Image Enhancement Pipeline That Integrates An Improved Atmospheric Scattering Model (ASM) With Otsu-based Segmentation. The System First Converts The RGB Input To Grayscale, Then Applies Otsu’s Method To Segment Fog-dense Regions. This Segmentation Guides A Region-specific Inverse ASM Dehazing: We Estimate Atmospheric Light And Transmission Differently For Fog And Non-fog Areas To Avoid Global Artifacts. Each Color Channel Is Then Enhanced According To The Refined ASM And Recombined To Preserve Color Fidelity. The Method Is Evaluated On Synthetic And Real Foggy Images Using Standard Metrics: Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), And Mean Squared Error (MSE). Results Show That Segmenting Out Heavy-fog Regions Before Applying ASM Yields Clearer, More Natural Images Compared To Baseline Defogging. For Example, Our Approach Achieves Higher PSNR And SSIM (closer To 1) And Lower MSE Than Conventional Methods, Confirming Its Effectiveness. We Include Example MATLAB Code Illustrating Grayscale Conversion And RGB Reconstruction.

Author: Aravindh.S | Dhamodharan.S | Solai balaji.G | Veyilraja.S

Area of research: Electronics And Computer Engineering

Chronic Immobility Resulting From Conditions Such As Stroke, Paraplegia, And Musculoskeletal Disorders Causes Severe Complications Including Muscular Atrophy, Pressure Sores, Poor Circulation, Joint Stiffness, And Impaired Mental Health. This Underscores A Critical Need For Effective, Technology-driven Rehabilitation Solutions To Enhance Patient Recovery And Quality Of Life. This Project Presents The Development Of A Multifunctional Intelligent Exoskeleton Robot Designed To Assist Patients With Movement Impairments Including Paraplegia And Post-stroke Recovery. The System Promotes Rehabilitation By Enabling Natural Human Movement Patterns—such As Standing, Walking, And Basic Exercises—while Simultaneously Maintaining Proper Posture And Boosting Circulation To Prevent Complications Associated With Immobility. Incorporating Modern Robotics, Biomechanics, Real-time Sensor Data (including SpO2 And Pulse Monitoring), And IoT Technology, The Exoskeleton Adjusts Therapy Protocols Remotely Based On Individual Patient Needs. It Includes Automated Safety Mechanisms That Halt Therapy If Vital Signs Fall Below Safe Thresholds, Ensuring Patient Protection. Additional Features Such As Integrated Massage And Heat Pads Assist In Muscle Relaxation And Circulation Improvement During Prolonged Therapy Sessions. The Proposed Solution Offers Cost-effective And Scalable Rehabilitation Support, Significantly Improving Functional Independence, Mobility, And Overall Patient Outcomes. This Intelligent Rehabilitation System Represents A Promising Advancement In Personalized, Safe, And Remotely Monitored Musculoskeletal Healthcare.

Author: Suraj Chavan | Samarth Kadolkar | Santosh Anuse | Ajinkya Killedar | Dhanashri Biradar

Area of research: Machine Learning & Deep Learning

Predicting Stock Prices Accurately Is Essential For Making Well-informed Decisions In Erratic Financial Markets. This Paper Introduces A Compositional Deep Learning Framework For Multivariate Time-series Forecasting That Integrates Three RNN Variants: LSTM, GRU, And SRU. Grey Wolf Optimizer (GWO) And Random Search (RS) Were Used To Develop And Optimize A Total Of 54 Model Architectures. The Best Results Were Obtained By LSTM-GWO (1-1-0-1), With R2 = 99.2427%, MAPE = 1.1721%, RMSE = 339.3902, WI = 0.9981, NSE = 0.9924, And Minimal Bias (PBIAS =0.0523). Additionally, GRU-GWO And SRU-GWO Performed Better Than RS-based Models, Demonstrating The Efficacy Of Metaheuristic Optimization. The Results Show That GWO And Systematic Architectural Design Greatly Improve Forecasting Accuracy And Model Stability For Reliable Financial Prediction Systems.

Author: Bharani K | Krishna Kumar R | Surendharan R | Dalphin Mary F

Area of research: Civil Engineering

In Many Parts Of The World, Especially In Developing Nations, Groundwater Is An Essential Source Of Agricultural And Drinking Water. However, The Quality Of Groundwater Has Drastically Declined Due To Fast Industrialization, Urbanization, And Agricultural Activity. Some Of The Main Reasons Of Its Contamination Are Residential And Commercial Waste, Landfills, Inadequate Drainage Systems, Seepage Via Sewage Lines, Agricultural Practices Including Heavy Fertilizer And Pesticide Usage, Irregular Rainfall, And Poor Management. In Order To Determine Whether Water Is Suitable For Home, Agricultural, And Industrial Applications, This Review Study Examines Groundwater Quality Using Important Physicochemical Characteristics. In Connection With The Water Quality Standards Set By The Bureau Of Indian Standards, Significant Parameters Like PH, Total Dissolved Solids (TDS), Hardness, Alkalinity, Dissolved Oxygen (DO), Biochemical Oxygen Demand (BOD), Chemical Oxygen Demand (COD), Nitrates, Chlorides, Sulphates, And Heavy Metals Are Examined.

Author: Ms. Yadav Sonam Rajpal | Indresh Kumar | Jyoti Sharma | Bushra Hashme | Sagar Gautam

Area of research: Management

Work-life Balance Is A Critical Issue For Women Entrepreneurs, Especially In Developing Countries Like India. This Study Analyses The Challenges Faced By Women Entrepreneurs In Balancing Their Professional And Personal Responsibilities Using Secondary Data. It Also Explores Influencing Factors Such As Family Support, Financial Pressure, And Time Management. The Study Highlights Strategies For Improving Work-life Balance And Presents Real Case Studies From India.

Author: P.Sowmiya | Dr.R.Rathidevi

Area of research: Mechanical Engineering

A Grass Cutter Machine Is A Mechanical Device Designed To Trim And Maintain Grass At A Uniform Height Efficiently. It Is Widely Used In Gardens, Agricultural Fields, Lawns, And Parks To Reduce Manual Effort And Save Time. The Machine Typically Consists Of A Motor Or Engine, Cutting Blades, A Supporting Frame, And A Handle For Operation. Depending On The Design, It Can Be Powered By Electricity, Battery, Or Fuel. The Rotating Blades Cut Grass Evenly, Improving The Appearance Of The Area And Promoting Healthy Growth. The Development Of Grass Cutter Machines Has Significantly Increased Productivity And Reduced Labor-intensive Work. Modern Machines Are Designed To Be Lightweight, Portable, And Easy To Operate, Making Them Suitable For Both Small-scale And Large-scale Applications. Some Advanced Models Also Include Adjustable Cutting Heights And Safety Features. Overall, The Grass Cutter Machine Is An Essential Tool In Landscaping And Agricultural Maintenance, Offering Efficiency, Precision, And Convenience.

Author: Shahid Mulani | Suleman Sayed | Shreyash Jadhav | Ritesh Kumar | Mr. Kadam G. J.

Area of research: Computer Science And Engineering

The Rapid Digitalization Of Academic Administration Has Become A Cornerstone Of High-performance Educational Institutions Worldwide. Traditional Manual Result Processing And Resource Management Systems Are Frequently Plagued By Mathematical Discrepancies, Security Vulnerabilities, And Delayed Feedback Loops That Hinder Institutional Efficiency. This Paper Presents SRMS-Ai, A Comprehensive Student Result Management System Designed To Bridge The Critical Gap Between Administrative Oversight And Student Accessibility. Built On The MERN Stack—MongoDB, Express.js, React.js, And Node.js—the System Integrates Advanced Security Through FIDO2/WebAuthn Biometric Authentication And Real-time Academic Performance Tracking Based On Anna University R-2021 Regulations. The Proposed Solution Employs A Robust Role-Based Access Control (RBAC) Mechanism To Facilitate Cross-departmental Collaboration Among Four Primary User Categories: Administrators, Heads Of Department (HODs), Faculty, And Students. Each Role Is Assigned Distinct Permissions And Workflows To Ensure Data Integrity And Operational Clarity. Preliminary Evaluations Indicate A Significant Reduction In Administrative Overhead, An Elimination Of Manual GPA Calculation Errors, And A Marked Increase In Data Integrity For High-stakes Academic Records. The System's Mobile-first Design Ensures Accessibility Across A Wide Range Of Devices, Addressing A Critical Gap In Existing Legacy Portals.

Author: Mr. G. Vadivel Murugan | M. R. Kamalesh | K. Dharun | R. Kumar

Area of research: Artificial Intelligence & Web Technologies

The Rapid Shift To Remote Hiring Has Created New Avenues For Interview Fraud. This Paper Presents An Automated AI System For Interview Fraud Analysis And Alerts, A Browser-native Web Application That Monitors Online Interview Sessions In Real Time Using A Multi-modal Detection Pipeline. The System Integrates Computer Vision Via MediaPipe Face Mesh And TensorFlow.js COCO-SSD For Face Monitoring, Eye Gaze Estimation, And Object Detection; Web Audio API For Multi-voice And Whisper Detection; A QR Code-based Mobile Device Pairing Module Backed By Firebase Realtime Database; And Browser Tab-switch Detection Using The Page Visibility API. All Violations Are Mapped To A Dynamic Integrity Score Beginning At 100 Points. An Administrator Dashboard Provides Post-session Review With Candidate-level Integrity Scores And Violation Histories. The System Extends Prior Work On Audio-visual Synchronization-based Fraud Detection By Delivering Proactive, Real-time Detection Across Six Independent Modalities Entirely Within The Candidate's Browser.

Author: B. Nirmal | P. Abinesh | A. Manikandan | Mrs. K. Menaka

Area of research: Computer Science And Engineering

With The Rapid Growth Of Digital Media And Social Networking Platforms, The Spread Of Misinformation And Fake News Has Become One Of The Most Pressing Challenges Of The Modern Era. Platforms Such As WhatsApp, Facebook, And Twitter Deliver Billions Of Messages Daily, A Significant Proportion Of Which Contain Fabricated, Manipulated, Or Misleading Content. Human Fact-checkers Cannot Scale To Address This Volume. This Paper Presents FakeFinder, A Context-aware News Credibility Detection System That Uses Natural Language Processing (NLP) And Machine Learning (ML) To Automatically Classify News Articles And Social Media Messages As Fake Or Real With An Associated Confidence Percentage. The System Implements A Complete NLP Preprocessing Pipeline — Text Cleaning, Tokenization, Stopword Removal, Lemmatization, And Custom Feature Engineering — Combined With TF-IDF Vectorization And Five ML Classifiers. Custom Features Including FEAT_HIGH_CAPS, FEAT_MANY_EXCLAIM, FEAT_FORWARD_MESSAGE, And FEAT_CREDIBLE_SOURCE Are Engineered To Capture The Distinct Linguistic Fingerprint Of Fake Content. The Best-performing Model Achieves Approximately 95% Accuracy. The System Is Deployed As A Flask-based Web Application With A REST API, Enabling Real-time Fake News Detection For Any Input Text. Experimental Results Confirm That FakeFinder Effectively Identifies Misinformation And Provides Reliable, Explainable Classification With High Accuracy

Author: C. Jeeva | G. Hariprasath | M. Rajkumar | Dr. S. Muthukumar

Area of research: Computer Applications

Agricultural Trading Systems Often Lack Transparency And Are Heavily Dependent On Intermediaries, Leading To Unfair Pricing And Inefficiencies. Farmers Frequently Struggle To Obtain A Fair Price For Their Produce, While Buyers Face Inconsistent Pricing And Delays. This Project Introduces A Secure And Transparent Double Auction Platform Designed Specifically For Fresh Agricultural Commodities. The System Allows Farmers And Buyers To Interact Directly Through An Online Marketplace. A Double Auction Mechanism Is Used To Determine Fair Prices Based On Real-time Supply And Demand. To Ensure Security And Trust, Blockchain Technology And Cryptographic Methods Are Incorporated. These Technologies Help Maintain Tamper-proof Records And Protect Transaction Data. Overall, The System Minimizes The Role Of Intermediaries, Enhances Pricing Transparency, And Improves The Efficiency Of Agricultural Trading.

Author: Arthi M | Senthamaraiselvi

Area of research: Commerce Computer Application

Human Resources Are The Most Valuable Asset Of Any Organization, And Their Performance Plays A Vital Role In Achieving Organizational Success. A Performance Management System (PMS) Helps Organizations Plan, Monitor, And Evaluate Employee Performance While Providing Feedback And Motivation To Improve Productivity. This Study Focuses On Examining The Effectiveness Of The Performance Management System And Its Influence On Employee Productivity At Aviram Knitters In Tirupur. The Study Analyses Employees’ Perceptions Of The Existing Performance Evaluation Process And How It Contributes To Their Work Efficiency And Motivation. Data Were Collected From Employees Using A Structured Questionnaire And Analysed Using Appropriate Tools. The Findings Help Identify The Strengths And Weaknesses Of The Current System And Provide Suggestions To Improve The Performance Management Practices For Enhancing Employee Productivity And Organizational Performance.

Author: Dr. M. PRAKASH | Mr.K. PRASANNA

Area of research: PSYCHOLOGY

This Research Examines How Family Structure—specifically, Whether A Young Adult Grew Up In A Single-parent Or Two-parent Household—affects The Way They Form Romantic Attachments. Using The Experiences In Close Relationships-Revised (ECR-R) Scale And The Parental Bonding Instrument (PBI), Data Were Collected From A Targeted Group Of 200 Young Adults. The Study Looks Into Whether Early Caregiving Experiences Are Connected To How Adults Feel About Trust And Intimacy In Relationships. The Findings Show A Weak, Non-significant Link (r = 0.066, P = 0.354) Between How Parents Bonded With Their Children And The Romantic Attachment Styles Of Adults. Additionally, No Significant Differences Were Found In Attachment Scores Based Solely On Family Structure.

Author: Dhana Sree M H | Mr. Manoj R | Ms. Nivedha | Nivedha K

Area of research: Psychology

Burnout Has Become A Critical Concern Among Healthcare Professionals, Particularly Community Pharmacists Who Are Exposed To Continuous Occupational Stressors Such As Heavy Workload, Long Working Hours, Administrative Responsibilities, And Frequent Patient Interactions. These Stressors Contribute To Psychological Burnout, Characterized By Emotional Exhaustion, Depersonalization, And Reduced Personal Accomplishment. The Present Study Aimed To Examine The Relationship Between Family Cohesion And Burnout Among Community Pharmacists. A Quantitative Correlational Research Design Was Employed, With A Sample Of 150 Community Pharmacists Selected Using Purposive Sampling. Data Were Collected Using The Family Adaptability And Cohesion Evaluation Scale (FACES IV) And The Maslach Burnout Inventory (MBI). Pearson’s Correlation Were Used For Data Analysis. The Findings Revealed A Statistically Significant Negative Correlation Between Family Cohesion And Burnout (r = -0.438, P < 0.01), Indicating That Higher Levels Of Family Cohesion Are Associated With Lower Levels Of Burnout. The Study Highlights The Importance Of Family Support As A Protective Factor In Reducing Occupational Stress And Promoting Psychological Well-being Among Pharmacists. These Findings Suggest The Need For Holistic Interventions That Incorporate Both Workplace And Family-based Support Systems.

Author: Prathiba K | Mr. Manoj R | Ms. Nivedha R

Area of research: NA

Normally, The People Who Are Physically Disabled Due To Many Reasons Such As Accidents, Severe Injuries, Paralysis Is Needed The Wheelchairs By Which They Will Be Able To Move Around. But It Is Difficult For The Physically Disabled People To Operate The Wheelchair. For Those Peoples This Voice Controlled Wheelchair Is Introduced. Wheelchairs Are Used When Handicapped People Needs To Travel Somewhere. But As We Said Physically Disabled Peoples Can’t Operate Wheelchair They Will Need Someone’s Help To Operate The Wheelchair. This Voice Controlled Wheelchair Will Help Physically Disabled Peoples To Travel Anywhere They Want Without Anyone’s Help By Just Giving Voice Commands. This Voice Controlled Wheelchair Will Also Help The Aged People Who Live Alone To Go Anywhere They Want Without Someone’s Help.

Author: Saujanya Sukhdev Pote | Aditi Dhanaraj Shinde | Parnavi Santosh Babar | Mokshda shivaji poul

Area of research: INFORMATION TECHNOLOGY

Managing Personal Finances Is A Critical Skill In Modern Life, Yet Many Individuals Struggle Due To A Lack Of Accessible And Intuitive Tools. This Paper Presents The Development Of A Comprehensive Personal Finance Tracker — A Web-based Application Aimed At Helping Users Monitor Income, Expenses, Savings, Budgets, And Financial Trends. Utilizing Modern Web Technologies Such As React, TypeScript, And Tailwind CSS, The Platform Delivers A Responsive And Interactive Experience. With Features Such As Budget Alerts, Dynamic Charts, Dark Mode Support, And Offline Persistence, This Application Addresses Common Challenges In Personal Finance Management. This Study Outlines The Development Approach, System Architecture, And Key Functionalities, And Proposes Future Improvements Based On User Feedback And Evolving Financial Needs.

Author: KARTHICK G | SHIVAN SELVAM | KUNNAL T | POOVARASAN A

Area of research: Mechanical Engineering

AeroVolt Is A Compact And Efficient Renewable Energy System Designed To Generate Electricity Using Wind Power Through A Vertical Axis Wind Turbine (VAWT). The Main Objective Of This Project Is To Utilize Low-speed Wind Energy Available In Urban And Rural Areas And Convert It Into Useful Electrical Energy In An Eco-friendly Manner. The System Consists Of Vertical Blades Mounted Around A Central Shaft, Which Rotate When Wind Flows From Any Direction. This Rotational Motion Is Transferred To A Generator, Where Mechanical Energy Is Converted Into Electrical Energy. Unlike Traditional Horizontal Wind Turbines, AeroVolt Does Not Require Alignment With Wind Direction, Making It More Suitable For Locations With Irregular Wind Patterns. The Design Is Simple, Cost-effective, And Requires Low Maintenance, Making It Ideal For Small-scale Applications Such As Street Lighting, Household Power Generation, And Educational Purposes. The AeroVolt System Contributes To Reducing Dependence On Fossil Fuels And Supports Sustainable Development By Promoting Clean And Green Energy. In Conclusion, AeroVolt Demonstrates An Innovative Approach To Harness Wind Energy Efficiently In A Compact Form, Making It A Promising Solution For Future Renewable Energy Needs.

Author: SavalsureShrvan Arvind | Jadhav Sanket Balasaheb | Gunjate Ashish vijay | Prof..Bidve M.A

Area of research: Computer Science And Engineering

The Rapid Growth Of Online Job Portals Has Changedhow People Find Work, Making The Process Much Easier. However, This Easy Access Has Also Led To A Massive Increase In Fake Job Postings And Employment Scams. Scammers Use These Platforms To Steal Personal Information And Trick Vulnerable Applicants Out Of Their Money. This Paper Proposes An Automated System To Detect And Identify Fake Job Ads On Popular Platforms Like LinkedIn, Naukri, Indeed, And Internshala. The System Uses Web Scraping Tools To Collect Live Job Data Directly From The Internet. Natural Language Processing (NLP) Techniques Are Then Used To Clean The Text And Find Suspicious Words Or Phrases. Finally, A Machine Learning (ML) Model Looks At These Features To Decide Whether The Job Is Real Or Fake. By Using This Automated Check, The System Gives Users A Safe And Highly Reliable Way To Verify Job Offers. Testing Shows That The System Can Successfully Catch Scams With Excellent Accuracy.

Author: Mrs.N.Sathiya Rani | M.Jeyakumar | B.Bala

Area of research: Computer

Modern Office IT Environments Are Highly Complex And Often Experience Unexpected System Failures That Impact Productivity And Increase Operational Costs. Traditional IT Support Systems Follow A Reactive Approach, Where Issues Are Addressed Only After Failures Occur, Leading To Delays And Inefficiencies. This Project Proposes A Dynamic IT Support System With Intelligent Triage And Operational Memory That Leverages Machine Learning Techniques To Proactively Detect And Classify System Failures. The System Continuously Monitors Key Performance Metrics Such As CPU Usage, Memory Utilization, Disk Activity, Network Latency, And Application Response Time. Using Predictive Models, The System Identifies Potential Failures Before They Occur And Classifies Them Into Categories Such As Hardware, Software, Network, Or Security Issues. Based On The Prediction, Automated Support Tickets Are Generated To Assist IT Personnel In Taking Preventive Action. The Proposed System Improves Fault Management Efficiency, Reduces Downtime, And Enhances System Reliability By Shifting From Reactive To Proactive IT Support.

Author: Teja Shree C V | Nathiya M | Ashish Christin C

Area of research: Computer Science And Engineering

Breast Cancer Recurrence Remains One Of The Most Critical Challenges In Long-term Oncology Care, Posing Significant Risks Even After Successful Primary Treatment. While Advances In Early Diagnosis And Therapeutic Interventions Have Improved Survival Rates, The Ability To Accurately Predict Recurrence Continues To Be Limited By Complex Biological Variability And Reliance On Subjective Clinical Judgment. This Paper Presents An Intelligent Breast Cancer Recurrence Prediction System Leveraging Machine Learning Techniques To Classify Patients Into Low-risk And High-risk Recurrence Groups. The Proposed System Integrates Clinical, Pathological, And Molecular Features—including Tumor Size, Lymph Node Involvement, Hormone Receptor Status, HER2 Expression, And Patient Demographics—within A Structured Data-driven Framework. Multiple Supervised Learning Algorithms Are Trained And Evaluated To Identify The Most Reliable Predictive Model. Experimental Results Demonstrate That Ensemble-based Classifiers Achieve Superior Performance In Terms Of Accuracy, Precision, Recall, And ROC-AUC. The System Aims To Support Oncologists In Personalized Treatment Planning, Proactive Monitoring, And Improved Post-therapy Decision-making, Thereby Enhancing Long-term Patient Outcomes In Clinical Practice.

Author: Ms. Suvitha S | Hasini P | Keerthana S | Nandhini K

Area of research: PSYCHOLOGY

Marriage Is A Complex Psychological And Social Institution Requiring Continuous Emotional Adjustment. This Study Examines The Relationship Between Self-regulation And Social Desirability Among Married Individuals. A Sample Of 150 Participants Aged 21–40 Years Was Selected Using Purposive Sampling. Data Were Collected Using The Self-Regulation Questionnaire (SSRQ) And The Marlowe-Crowne Social Desirability Scale. Statistical Analysis Revealed A Significant Negative Correlation (r = –0.278, P < .01), Indicating That Individuals With Higher Self-regulation Demonstrate Lower Socially Desirable Responding. The Findings Suggest That Stronger Internal Regulation Promotes Authentic Emotional Expression Rather Than Impression Management. The Study Contributes To Understanding Emotional Functioning Within Marital Relationships And Offers Implications For Counselling And Psychological Interventions.

Author: Harini S J | Mr. Manoj R | Ms. Kaviyapraba N

Area of research: Computer Science And Engineering

Traditional Data Warehouse Systems Require Users To Write Structured SQL Queries To Retrieve Information, Which Is Difficult For Non-technical Users. Moreover, Sensitive Organizational Data Stored In Warehouses Demands Controlled Access And Privacy Preservation. Existing Keyword Search Approaches Provide Basic Retrieval But Fail To Enforce Secure Role-based Data Visibility.This Paper Proposes A Privacy-Preserving Intelligent Keyword Search System Integrated With Role-Based Access Control (RBAC) For Cloud-based Data Warehouses. The System Converts Natural Language Keywords Into Structured Database Queries And Dynamically Filters Results Based On User Authorization Levels. Public Data Is Accessible To All Users While Confidential Records Remain Restricted To Administrators. The Proposed Model Improves Usability, Security, And Controlled Data Sharing In Enterprise Environments.

Author: Baalasubramani V | Shivanad N | Rohith B | Subramani V

Area of research: Computer Science And Engineering

Robotic Manipulators Are Extensively Used In Industrial Automation For Precision-driven Tasks Such As Object Handling, Assembly, And Material Transfer. Accurate Motion And Load Analysis Are Essential To Ensure Operational Stability And Prevent Mechanical Failure. This Paper Presents A Multibody Dynamic Simulation Framework For Evaluating The Motion Behaviour And Load-bearing Capacity Of A Robotic Manipulator Using PyBullet. A Multi-degree-of-freedom Robotic Arm Model Was Developed Using URDF With Realistic Mass And Inertia Properties. The System Integrates Forward Kinematics, Inverse Kinematics, And Newton–Euler Dynamic Modelling To Analyse Joint Torque Under Varying Payload Conditions. In Addition, The Framework Incorporates A Digital Twin–based Simulation Environment With A Graphical User Interface For Parameter Input And Control, Along With AI-based Object Detection Using The YOLO Model For Identifying Target Objects Within The Simulation. Real-time Simulation Data, Including Joint Angles, Torque Values, Positional Parameters, And End-effector Trajectories, Were Logged And Analyzed For Performance Evaluation. Experimental Results Demonstrate The Relationship Between Payload Mass And Torque Requirements, Identifying Safe Operational Thresholds. The Developed System Was Successfully Implemented And Tested, Demonstrating That Physics-based Digital Twin Simulation Provides An Efficient And Reliable Approach For Analysing Robotic Manipulator Dynamics Before Hardware Implementation.

Author: V S Sagarika | Thanmayee R | Prof. Dr. N.Saravanan | Prof. X. Anitha Sarafin

Area of research: Computer Science

This Paper Presents An Artificial Intelligence Driven Project Management System Designed To Predict Project Delays And Assess Associated Risks Using A Hybrid Analytical Approach. The Proposed System Integrates Machine Learning And Natural Language Processing To Evaluate Both Structured Project Data And Unstructured Textual Updates. A Random Forest Model Is Employed To Estimate Delay Probability Based On Key Project Attributes, While A Rule Based Text Analysis Module Identifies Risk Indicators From Progress Reports And Team Communications. These Outputs Are Further Combined Through A Risk Fusion Mechanism To Generate A Comprehensive Risk Score, Enabling More Accurate And Context Aware Decision Making. In Addition, A Root Cause Analysis Component Is Incorporated To Provide Interpretable Insights Into The Factors Contributing To Potential Delays, Thereby Supporting Proactive Intervention Strategies. The System Is Implemented Using A Flask Based Backend With A Lightweight Database For Data Management And A User Interface For Interaction. Experimental Evaluation On An Agile Project Dataset Demonstrates The Effectiveness Of The Integrated Approach In Identifying Risk Patterns And Improving Early Detection Of Delays. The Proposed Framework Offers A Practical And Scalable Solution For Enhancing Project Monitoring, Reducing Uncertainty, And Supporting Informed Managerial Decisions In Dynamic Development Environments.

Author: T. Subha Rathi Priya | Mirthika S | Linga praba G

Area of research: Computer Science And Engineering

The Approach Extracts Key Features From Web Documents To Form Concepts, Which Are Used To Build An Ontology Capturing Semantic Relationships Between Learning Materials. Web Usage Mining Analyzes Learner Interactions And Navigation Patterns, While Semantic Clustering Groups Content Based On Similarity To Identify Learner Preferences And Difficulties. By Combining Semantic Knowledge With Usage Data, The System Provides Personalized And Context-aware Recommendations Tailored To Each Learner’s Understanding. Experimental Results Show That The System Effectively Supports Learner Review By Recommending Relevant Content Aligned With Individual Needs. It Helps Learners Focus On Weak Areas, Improves Engagement, Reduces Cognitive Overload, And Enhances Overall Comprehension. This Framework Demonstrates The Effectiveness Of Integrating Ontology With Web Usage Mining For Adaptive Learning Systems.

Author: Mrs. P. Nirmala | Mr. M. Mutharasu | Mr. K. Kaviarasu | Mr. Z. Mohammed Nazrudeen

Area of research: Computer Science And Engineering

Energy Sector Stock Price Prediction Is A Critical Yet Challenging Task In Financial Forecasting, Characterized By High Volatility, Non-linearity, And Complex Interdependencies Driven By Geopolitical Events, Regulatory Shifts, And Commodity Price Fluctuations. Traditional Statistical Models And Standalone Deep Learning Approaches Fail To Simultaneously Capture Both The Temporal Dynamics Within Individual Stock Sequences And The Spatial Dependencies That Exist Between Correlated Companies. This Paper Proposes An Intelligent Hybrid GCN-LSTM Model Augmented With The Relative Strength Index (RSI) As A Domain-specific Technical Momentum Indicator. The Proposed Architecture Integrates A Graph Convolutional Network (GCN), Which Employs Dynamic Time Warping (DTW)-based Correlation Analysis To Construct An Inter-stock Graph Capturing Spatial Dependencies, With A Long Short-Term Memory (LSTM) Network Enhanced By An Attention Mechanism For Modelling Temporal Price Patterns. RSI Is Incorporated As An Additional Input Feature, Providing Overbought And Oversold Market Signals That Enrich The Model's Understanding Of Momentum-driven Price Reversals. Comprehensive Experiments Conducted On 30 Top Global Energy Sector Stocks Sourced From Yahoo Finance, Covering The Period From March 1, 2011 To September 26, 2024, Demonstrate That The Proposed GCN-LSTM+RSI Model Achieves A Mean Squared Error (MSE) Of 0.061, Root Mean Squared Error (RMSE) Of 0.247, Mean Absolute Error (MAE) Of 0.198, And A Coefficient Of Determination (R²) Of 0.872. These Results Represent A 21.8% Improvement In MSE Over The Base GCN-LSTM Model And Significantly Outperform Standalone LSTM, GRU, MLP, And Linear Regression Baselines.

Author: Mrs. Banuppriya P | Yokesh Kumar E | Ranjith Kumar S | Poovarasan M | Arun Kumar R

Area of research: Mechanical Engineering

The Sand Sieving Machine Is A Mechanical Device Designed To Separate Sand Particles Based On Their Size And Remove Unwanted Materials Such As Stones, Gravel, And Impurities. In Traditional Methods, Sand Sieving Is Done Manually Using Hand Sieves, Which Requires More Time, Labor, And Physical Effort. The Proposed Sand Sieving Machine Uses An Electric Motor, Crank And Slider Mechanism (or Rotary Drum), Shaft, Bearings, And Mesh Screen To Perform The Sieving Operation Efficiently. The Machine Works By Vibrating Or Rotating The Sieve, Allowing Fine Sand To Pass Through The Mesh While Larger Particles Remain On The Screen. This Machine Improves Productivity, Reduces Human Effort, And Ensures Uniform Sand Quality. It Is Simple In Construction, Cost-effective, And Suitable For Construction Sites And Small-scale Industries

Author: AutiRohit Sambhaji | Bhalke Ramji Vyankat2, | Shinde Dhiraj Madhav | Wadekar Shrishailya Umakant | Prof.Dharashive G.M. | Prof.Jadhav P.C. | Prof. Nitalikar P.S.

Area of research: MANUFACTURING ND RESEARCH

The Sugarcane Lifting Machine Is A Mechanical Device Designed To Reduce Manual Labor And Improve Efficiency In Sugarcane Handling During Harvesting And Loading. In Traditional Methods, Workers Manually Lift And Load Heavy Bundles Of Sugarcane, Which Requires More Time, Labor, And Physical Effort. The Proposed Sugarcane Lifting Machine Uses A 1 HP Gear Motor, Chain And Sprocket Mechanism, Hollow Shaft, Bearings, And A 60-feet Square Rod Frame To Lift And Move Sugarcane From Ground Level To A Higher Platform Or Vehicle. The Machine Operates Using A Motor-driven Chain Conveyor System That Lifts The Sugarcane Smoothly And Continuously. Bearings Are Used To Reduce Friction And Ensure Smooth Rotation Of The Shaft, While The Sprocket And Chain Transmit Power Efficiently. This Design Helps Increase Productivity, Reduce Worker Fatigue, And Minimize Time Required For Loading Sugarcane. The Sugarcane Lifting Machine Is Simple In Construction, Cost-effective, And Easy To Operate, Making It Suitable For Small And Medium-scale Farmers. It Improves The Overall Harvesting Process By Providing A Faster, Safer, And More Efficient Method Of Lifting And Transporting Sugarcane.

Author: Chavan Aditya Bhanudas | Jivlage Bhavan Angad | Pawar Ganesh Madhukar | Ratnapurkar Saksham Sanjay | Prof.Dharashive G.M. | Prof. Jadhav P.C

Area of research: Computer Science And Engineering

Chronic Kidney Disease (CKD) Is A Non-communicable Illness That Affects A Significant Portion Of The Global Population. Major Risk Factors Include Diabetes, Hypertension, And Cardiovascular Disorders. CKD Often Remains Asymptomatic In Its Early Stages, Leading To Delayed Diagnosis And Potentially Fatal Outcomes. This Project Proposes A Machine Learning-based Approach For Early CKD Prediction Using Ensemble Algorithms. Four Ensemble Models—Random Forest, Gradient Boosting, Bagging, And AdaBoost—are Implemented To Diagnose CKD At An Early Stage. The Models Are Evaluated Using Multiple Performance Metrics, Including Accuracy, Sensitivity, Specificity, Precision, F1-Score, Mathew Correlation Coefficient (MCC), And Area Under The Curve (AUC). Experimental Results Indicate That The Random Forest Model Outperforms Other Algorithms, Achieving The Highest Accuracy, Sensitivity, Precision, MCC, And AUC Scores. The Proposed System Demonstrates The Potential To Assist Medical Practitioners In Early CKD Detection, Enabling Timely Intervention And Improving Patient Outcomes.

Author: Mrs.P.Kavitha Pandian | Ms. A. Soundarya | Ms.S.Manjuladevi | Ms.P. Madhumitha

Area of research: CYBER SECURITY

In The Modern Digital Era, Secure Communication Has Become A Critical Requirement Due To The Rapid Growth Of Cyber Threats, Data Breaches, And Unauthorized Access To Confidential Information. This Paper Presents A Highly Secure Web-based Communication System Developed Using Python Flask And MySQL That Integrates Advanced Cryptographic And Steganographic Techniques To Ensure Multi-layer Protection Of Transmitted Data. The System Enables Authenticated Users To Register, Log In, And Securely Exchange Messages In The Form Of Text Or Files. Unlike Traditional Messaging Systems That Rely Solely On Encryption, This Application Combines AES-GCM (Advanced Encryption Standard – Galois/Counter Mode) Encryption With LSB (Least Significant Bit) Image Steganography To Provide Double-layer Security. In The Proposed System, The User First Encrypts The Message Payload Using AES-GCM Encryption With A Secret Passphrase. AES-GCM Ensures Confidentiality, Integrity, And Authentication By Generating Secure Cryptographic Components Such As Salt, Nonce, Ciphertext, And Authentication Tag. The Encrypted Payload Is Then Embedded Inside A Cover Image Using LSB Steganography, Which Hides The Existence Of The Secret Data By Modifying The Least Significant Bits Of Image Pixels Without Visibly Altering Image Quality. After Embedding, The Generated Stego Image Is Again Encrypted Using AES-GCM Before Being Stored Or Transmitted, Thereby Adding An Additional Security Layer. This Dual Protection Mechanism Prevents Attackers From Detecting Hidden Data Even If The Encrypted File Is Intercepted. Experimental Results Demonstrate That The Proposed System Achieves High Imperceptibility With PSNR Values Exceeding 60 DB And Provides Robust Protection Against Unauthorized Access And Detection Attacks.

Author: Gayathri K | Keshava Varshini M | Vani Shree V | Mr. S. Ganeshkumar

Area of research: Computer Science And Engineering

The Increasing Prevalence Of Stress, Anxiety, And Depression Among Students Has Highlighted The Need For Accessible, Personalized, And Technology-driven Mental Health Support. This Project Proposes Mind Space, A Comprehensive Digital Platform Designed To Enhance Student Well-being Through Intelligent Monitoring, Assessment, And Intervention. At Its Core, The Platform Incorporates A Depression Analysis System That Leverages Artificial Intelligence (AI), Natural Language Processing (NLP), And Sentiment Analysis To Engage Users In Empathetic Conversations, Detect Linguistic Patterns Indicative Of Depression, And Provide Ongoing Assessments For Timely Intervention. The System Adapts Follow-ups And Recommendations Based On Symptom Severity, Enabling Personalized Mental Health Support. In Addition, The Platform Offers Mood Tracking, Self-assessment Tools, Anonymous Peer Support, Guided Mindfulness Exercises, And Access To Professional Counseling, Bridging The Gap Between Mental Health Needs And Resource Accessibility. Experimental Evaluation Demonstrates The Platform’s Effectiveness In Early Detection Of Depressive Symptoms, Enhancing Mental Health Care Accessibility, And Supporting Proactive Well-being Management Among Students.

Author: Mr.G.Vadivelmurugan | Ms.S.Yuvapriya | Ms.A.Iswarya

Area of research: Computer Science And Engineering

Digital Platform Of Combined Agriculture Trading: E-NAM, E-RaKAM Has Make Online Selling And Bidding Possible For Agriculture Products. But Most Farmers Find It Hard To Operate These Platforms Because Of The Complex Interfaces, Little Instructions And Less Marketing Experience. In This Paper, We Present Agri Reach AI, An Intelligent Agricultural Trading Platform That Combines Voice Input, Chatbot Assistance With AI-based Goods Detection And Transaction Tracking For Easier Interaction To Farmers In The Digital Marketplace. The Web System Provides A Simple Interface For The Farmers To Submit Their Product Details, Identify Products, Place Bids And Track Transaction History Transparently. Moreover, The Proposed System Improves Online Agricultural Trading Usability, As Well As The Trust And Participation Of Farmers By Integrating Artificial Intelligence Into Native Language Usage Interaction.

Author: Mrs. J. B. Pradeepa | M Akshaya | S Gayathri | R Jayasri | G Mahalakshmi

Area of research: MACHINE LEARNING

Rapid Urban Growth Has Intensified Environmental Degradation, Particularly In The Form Of Declining Air Quality And Shrinking Green Spaces. Unregulated Development, Population Density Increases, And Industrial Expansion Have Disrupted Ecological Balance And Elevated Atmospheric Pollution Levels. To Address These Challenges, This Study Presents An AI-driven Geospatial Simulation Framework Designed To Analyze Urban Growth Dynamics And Forecast Pollution Severity. The Model Integrates Satellite Imagery, GIS-based Datasets, And Multiple Environmental Parameters To Examine Spatial Transformations In Metropolitan Areas. A Hybrid Ensemble Strategy Combining XGBoost And AdaBoost Is Utilized To Enhance Predictive Robustness And Model Efficiency. The Framework Incorporates Both Spatial And Temporal Features To Categorize City Zones According To Pollution Intensity. Additionally, A Flask-based Backend Supports Real-time Analysis And Interactive User Access. The System Identifies Vulnerable Pollution Hotspots, Projects Future Environmental Risks, And Visualizes Outcomes Using Geospatial Mapping Techniques. By Enabling Evidence-based Urban Planning And Environmental Policy Decisions, The Proposed Framework Provides A Scalable And Intelligent Solution For Sustainable City Development.

Author: Dr.J.Paramesh | Sunayana G | Kalpana S U | Srihitha L

Area of research: Computer Science And Engineering

We Spent A Fair Amount Of Time Watching How Recruiters Actually Screen Resumes Before Building This. What Struck Us Was Not The Speed — Six To Eight Seconds Per Resume Is Well-documented [1] — But How Much Of That Time Went To Things That Had Nothing To Do With Whether The Candidate Could Do The Job. Things Like Font Choices, Gap Years, University Names. That Observation Is What Pushed Us Toward Building This. The System Pairs An NLP Extraction Layer With Five ML Models Running In Parallel. Logistic Regression Handles ATS Compatibility Scoring And Lands At 83%, Random Forest Picks Up Job Role Detection At 92%, Gradient Boosting Estimates Experience Level At 90%, Isolation Forest Catches Suspicious Submissions At 88%, And SVM Handles Quality Tiering At 86%. Eight Field Types Get Pulled From PDF And DOCX Files With 85 To 98 Percent Accuracy Depending On The Field. The Whole Pipeline Wraps Up In Around 6.2 Seconds — Which In Practice Cuts First- Pass Screening Time By About 65% Compared To Doing It By Hand.

Author: Mr. Dinesh | B Ajmal Shakeel | S Gowthamraj | K H Hamdan Mohammed | R Haynesh

Area of research: Information Technology

Road Accidents Caused By Drunk Driving And Sudden Driver Health Issues Are A Serious Global Concern. Many Accidents Occur When Drivers Operate Vehicles Under The Influence Of Alcohol Or When They Experience Unexpected Medical Conditions Such As Dizziness, Fatigue, Or Unconsciousness While Driving. These Situations Can Lead To Loss Of Vehicle Control And Severe Accidents. To Address This Issue, This Project Proposes An Intelligent Alcohol Detection And Emergency Vehicle Control System That Improves Vehicle Safety And Helps Prevent Potential Road Accidents. The System Uses An Alcohol Sensor To Detect The Presence Of Alcohol In The Driver's Breath Before The Vehicle Starts. If The Detected Alcohol Level Exceeds The Permissible Limit, The System Automatically Prevents The Vehicle From Starting, Thereby Reducing The Risk Of Drunk Driving. In Addition, The System Incorporates An Emergency Safety Mechanism To Handle Situations Where The Driver Feels Unwell While Driving. An Emergency Button Placed Near The Steering Wheel Allows The Driver To Alert The System In Case Of A Sudden Health Issue. Once The Button Is Pressed, The Control Unit Activates Safety Sensors To Monitor The Vehicle’s Surroundings And Gradually Brings The Vehicle To A Safe Stop. This Controlled Stopping Mechanism Helps Avoid Sudden Accidents And Ensures The Safety Of The Driver, Passengers, And Pedestrians. Overall, The Proposed System Integrates Alcohol Detection Technology With An Emergency Vehicle Control Mechanism To Enhance Road Safety. It Provides An Effective Solution To Reduce Accidents Caused By Impaired Driving And Unexpected Driver Health Emergencies, Thereby Contributing To Safer And Smarter Transportation Systems

Author: Mr.B Sarvesan | Priyadharshini L | Reshma Shaik | Priyadharshini S

Area of research: Public Administration

Scientific Management Represents One Of The Earliest Systematic Attempts To Apply Scientific Principles To Organizational Management. Developed By Frederick Winslow Taylor In The Early Twentieth Century, The Theory Sought To Enhance Industrial Efficiency Through The Application Of Systematic Observation, Measurement, And Standardization Of Work Processes. This Article Examines The Conceptual Foundations Of Scientific Management And Analyses Its Major Principles And Techniques, Including Time And Motion Studies, Functional Foremanship, And The Differential Piece-rate System. The Study Also Evaluates The Theoretical Contributions Of Taylor In Transforming Management From A Traditional Rule-of-thumb Practice Into A More Structured And Scientific Discipline. At The Same Time, The Article Critically Explores The Limitations Of Scientific Management, Particularly Its Mechanistic View Of Workers And Its Neglect Of Social And Psychological Factors Within Organizations. By Reviewing Classical And Contemporary Literature, The Paper Highlights The Enduring Relevance Of Scientific Management In Modern Organizational Practices While Acknowledging The Need For More Human-centred Approaches In Management Theory.

Author: Goutam Kumar Shaw

Area of research: Pharmacy

The Goal Of This Work Is To Create A Herbal Face Wash Using Natural Herbs Such As Neem, Orange, Tulsi, Rose, Honey, Almond Oil, Aloe Vera Gel, Lemon Grass, Banana Peel, Rise Flour, And Linseed. The Resulting Face Wash Was Evaluated For Colour, Consistency, Ph, Washability, Homogeneity, Viscosity, Stability Studies, Grittiness, And Skin Irritability. And The Outcomes Were Consistent With The Face Cleansers On The Market; The Specifics Are Covered In The Study. Natural Medicines Are More Acceptable Because People Believe They Are Safer And Have Fewer Adverse Effects Than Manufactured Ones. Herbal Formulations Are Becoming Increasingly Popular In The Global Market. The Current Study Focusses On The Formulation And Evaluation Of A Herbal Anti-acne Face Wash Containing An Aqueous Extract Of Neem Leaves (Azadirachta Indica), Turmeric (Curcuma Longa), Liquorice Root, Shahi Jeera, Orange Peel, And A Hydroalcoholic Extract Of Nutmeg Fruit (Myristica Aroma). Although There Are Several Topical Herbal Formulations For Acne On The Market, We Recommend Creating A Pure Herbal Formulation That Contains No Synthetic Ingredients. The Plants Have Been Shown In The Literature To Have Good Antimicrobial, Antioxidant, And Anti- Inflammatory Properties. It Was An Excellent Attempt To Develop A Natural Anti-acne Face Wash Containing An Aqueous Extract Of Neem Leaves And Turmeric.

Author: Tanaya Hiralal Gedam | Nutan Khemraj Pustode

Area of research: Computer Science And Engineering

With The Rapid Growth Of Social Media Platforms, Unauthorized Sharing And Misuse Of Personal Images Has Become A Major Privacy Concern. Many Users Upload Photos Containing Other Individuals Without Their Permission, Which May Lead To Privacy Violations And Identity Misuse. This Paper Proposes A Secure Image Publishing System Using Facial Identity Verification That Detects Faces In Uploaded Images And Verifies Whether The Person In The Image Has Granted Permission For The Upload. The System Uses Computer Vision Techniques To Detect Faces And Compare Them With Registered User Data Stored In A Secure Database. If A Match Is Found, The System Sends A Notification To The Identified Person And Requests Approval Before Allowing The Image To Be Published. The Proposed System Integrates Technologies Such As Python, OpenCV, And Machine Learning–based Face Recognition To Ensure Accurate Identification And Permission Control. By Introducing An Automated Permission Mechanism, The System Helps Prevent Unauthorized Photo Sharing And Enhances Privacy Protection On Social Media Platforms. Experimental Results Demonstrate That The System Can Effectively Detect Faces And Manage Image Ownership Permissions With Reliable Performance.

Author: Mrs. N. Sujithaa | U. Jeevarathinam | N.R.S. Ramana | S. Jeyasakthi

Area of research: Pharmacy

There Has Been A Surge In Interest In Using Ayurvedic Medicines In Recent Years. Since Ancient Times, People Have Employed GlycyrrhizaGlabra (liquorice), ZingiberOfficinale (ginger), And Curcuma Longa (turmeric) As Medicines To Alleviate Coughs. Ayurveda Made Reference To The Usage Of These Herbal Medications. Cough Is A Prevalent Illness Problem That Affects People Of All Ages. Oral Medication Administration Is The Most Popular Method Due To Its Convenience Of Use, Fewer Sterility Requirements, Variable Dosage Form Design, And Improved Patient Compliance. The Objective Of This Research Project Is To Create Polyherbal Chewable Tablets Using The Wet Granulation Method For A Variety Of Ayurvedic Medications And Assess The Formulations For A Range Of Pharmaceutical Criteria. The Goal Of This Study Was To Create Chewable, Polyherbal Pills With Turmeric, Ginger, And Liquorice. These Chewable Polyherbal Tablets Were Created Using The Wet Granulation Method With A 5% W/v Acacia Gum Binder. The Final Polyherbal Chewable Tablet's Quality Was Assessed For Both Pre- And Post-formulation Parameters. The Preformulation Tests That Are Assessed For The Manufactured Powder Mixture (blend) Include Bulk And Tapped Densities, Hausner's Ratio, Carr's Index, And Angle Of Repose. General Look, Pill Size And Shape, Hardness, Friability, Weight Variation, And Disintegration Time Were All Evaluated For Polyherbal Chewable Tablets.

Author: Hemant Ramesh Neware | Nutan Khemraj Pustode

Area of research: Computer Science

The Rapid Developments In Smart Health Care Applications The Usage Of Machine Learning Models, Enhance The Quality Of Disease Prediction Process, The Outcome And The Accuracy. The Need For Classifying The Diseases From The Massive Data Collected From Health Care Infrastructure Is Difficult. The Datasets Are Not Similar And Structured. Various Challenges Faced By The Platform Users To Retrieve The Analysis Results Within The Time. The Need For Effective Processing Of Healthcare Data Is Demandable. The Proposed System Comprised Of Multiple Machine Learning Algorithm Comparisons To Evaluate The Performance Of Prediction Quality As Well As Classify The Electronic Health Care Records (EHR) Towards Various Disease Category. Leveraging The Predictive Capability Of The Proposed Model, The Comparison Towards Performance, Disease Formulation Is Evaluated. The Proposed System Is Implemented With Authorized Web Application Accessed By The Health Care Professionals, To Analyse The Diseases. The Quality Of Prediction Is Increased Towards The Quality Of Input Dataset And Preprocessing Quality.

Author: Vijayakumar M | Navinbharathi M

Area of research: Computer Science And Engineering

Modern Web Applications Are Increasingly Targeted By Automated Bots And Sophisticated Attackers Using Advanced Exploitation Techniques Such As Injection Attacks, Credential Stuffing, And Reconnaissance-based Probing. Traditional Intrusion Detection Systems Primarily Focus On Detection And Blocking, Often Failing To Extract Actionable Intelligence From Adversarial Interactions. This Paper Presents A Cloud-based, AI-powered Threat Deception Platform That Actively Engages Attackers Through Realistic Honeypot Interfaces And Tarpit Mechanisms While Simultaneously Analyzing Behavioral And Payload-level Data. The Proposed System Integrates Rule-based Attack Signature Detection With An XGBoost-based Behavioral Machine Learning Model To Identify Malicious Activity With High Accuracy. Severity Assessment Is Performed Using CVSS 3.1 Scoring, And Detected Threats Are Mapped To OWASP Top 10 Categories And Relevant CVE References. The Platform Is Fully Deployed On Cloud Infrastructure Using Firebase Hosting, A Flask-based Backend, And Azure Blob Storage For Scalable Logging. Experimental Evaluation Demonstrates Effective Detection Of Multiple Attack Vectors Including XSS, SQL Injection, Command Injection, And Automated Bot Behavior, While Maintaining Low Operational Cost. The Results Indicate That The Proposed System Not Only Detects Threats But Also Converts Attacks Into Valuable Security Intelligence.

Author: Mrs. P. Elakkiya | S Aakash | S Ahamed Asarudeen | S Kirthik Sarvash | S Kirthik Sarvash

Area of research: Mechanical Engineering

Water Scarcity Is Becoming A Major Global Concern Due To Rapid Urbanization, Industrial Growth, And Climate Change. At The Same Time, Large Amounts Of Wastewater Generated From Domestic, Commercial, And Industrial Sources Are Discharged Without Proper Reuse. This Project Focuses On The Design And Implementation Of An Efficient Filtration System To Treat Wastewater For Reuse In Air Conditioning (AC) Systems. Air Conditioning Units Require Significant Amounts Of Water For Cooling Towers, Condensers, And Heat Exchange Processes. Instead Of Using Fresh Potable Water, This Project Proposes The Use Of Treated Wastewater After Removing Impurities Such As Suspended Solids, Dissolved Salts, Microorganisms, And Organic Contaminants. The Treatment Process Includes Multiple Filtration Stages Such As Sediment Filtration, Activated Carbon Filtration, Membrane Filtration (RO/UF), And Optional UV Sterilization To Ensure Water Quality Meets Operational Standards. The System Is Designed To Reduce Freshwater Consumption, Minimize Environmental Pollution, And Lower Operational Costs. Water Quality Parameters Such As PH, Turbidity, Total Dissolved Solids (TDS), And Microbial Content Will Be Monitored Before And After Filtration To Evaluate Performance. This Project Demonstrates A Sustainable And Cost-effective Approach For Water Conservation In HVAC Systems And Promotes Environmentally Responsible Engineering Practices.

Author: Birajdar Krushna Shesherao | Garad Sachin Sambhaji | Pawar Pruthviraj Harishchandra | Bodhane Sanskruti Santosh | Prof.Bidve M.A

Area of research: Artificial Intelligence And Machine Learning

Autism Spectrum Disorder (ASD) Is A Neuro Developmental Condition Characterized By Challenges In Social Communication, Emotional Expression, And Behavioral Flexibility. Early Screening Plays A Crucial Role In Enabling Timely Intervention And Improving Developmental Outcomes In Children. However, Traditional Diagnostic Procedures Depend Heavily On Expert Observation And Standardized Clinical Assessments, Which Can Be Time-consuming And Inaccessible In Many Regions.. This Paper Presents A Multi-modal Autism Spectrum Disorder Detection System For Children That Integrates Behavioral Screening, Facial Emotion Recognition, And Speech Pattern Analysis Within A Unified Artificial Intelligence Framework. The System Employs A Random Forest Classifier For Questionnaire-based Behavioral Screening, A MobileNetV2 Deep Learning Model For Facial Emotion Detection, And A Machine Learning Speech Analysis Model For Identifying Atypical Vocal Characteristics. Each Modality Is Processed Through Dedicated Preprocessing And Feature Extraction Pipelines Before Being Integrated Through A Decision-level Fusion Mechanism To Generate The Final ASD Risk Prediction.. A Web-based Application Built Using The Flask Framework Enables Users To Submit Questionnaire Responses, Upload Facial Images, And Record Speech Samples. Experimental Evaluation Demonstrates That The Multi-modal Approach Improves Predictive Accuracy Compared To Single-modality Methods. The Proposed System Provides A Scalable, Accessible, And AI-assisted Screening Tool That Supports Caregivers And Clinicians In Early ASD Risk Identification.

Author: Ms. Deva Dharshini | Rangesh S | Dharmesh S | Abinash R

Area of research: Environmental Science

Microplastics Are Plastic Particles Smaller Than 5 Mm That Originate From The Degradation Of Larger Plastic Materials Or Are Manufactured For Commercial Use. In Recent Years, Microplastic Pollution Has Become An Important Environmental Concern Because These Particles Are Now Found In Water, Soil, And Even The Atmosphere. Due To Their Small Size, Microplastics Can Be Easily Ingested By Living Organisms And May Accumulate In Biological Systems. The Present Paper Reviews The Sources, Distribution, And Possible Impacts Of Microplastics On Living Organisms, Particularly Aquatic Species. It Also Discusses The Ecological And Potential Human Health Implications Associated With Microplastic Pollution

Author: Namisha Bagharia | Namisha Bagharia

Area of research: Gas Sensing Systems

The Advanced Gas Thermal Sensing System (AGTSS) Is An IoT-enabled Industrial Safety Monitoring Platform Designed To Detect Hazardous Gases And Thermal Anomalies In Real Time. Industrial Environments Such As Chemical Plants, Manufacturing Facilities, Oil Refineries, And Storage Units Are Highly Vulnerable To Gas Leaks And Temperature Fluctuations That May Lead To Severe Accidents, Environmental Damage, And Operational Disruptions. Traditional Monitoring Systems Rely On Standalone Sensors And Manual Inspections, Which Often Fail To Provide Timely Alerts And Centralized Monitoring. The System Utilizes Cloud-based Storage And Analytics, Enabling Real-time Monitoring Through Web Dashboards And Mobile Applications. Advanced Encryption Techniques Such As AES-256 Encryption And TLS Protocols Ensure Secure Transmission And Storage Of Sensor Data.

Author: Dr.S.Velmurugan | Thanigaivelavan A | Roshan N S

Area of research: Computer Engineering

The Rapid Growth Of Web Applications Has Led To An Increased Attack Surface For Cyberattacks Such As Structured Query Language (SQL) Injection, Cross-Site Scripting (XSS), And Other Application-layer Exploits. Traditional Web Application Firewalls (WAFs) That Rely Solely On Static, Signature-based Rules Struggle To Detect Obfuscated Payloads, Zero-day Attacks, And Novel Variants Of Existing Threats. This Paper Proposes An Intelligent Hybrid WAF Architecture That Combines Signature-based, Anomaly-based, And Machine Learning–based Detection To Provide Robust, Real-time Protection For Modern Web Applications. The System Monitors And Filters Hypertext Transfer Protocol (HTTP) Traffic Between Clients And The Web Application, Using A Multi-stage Detection Engine To Identify Malicious Requests And Apply Appropriate Mitigation Actions. The Proposed Model Leverages Public And Synthetic Web Attack Datasets For Training And Evaluation, With A Focus On SQLi And XSS Detection While Remaining Extensible To Other Emerging Threats. Expected Outcomes Include Improved Detection Accuracy, Reduced False Positives And False Negatives, Scalability In Cloud-native Environments, And A User-friendly Monitoring Dashboard That Supports Effective Security Operations.

Author: Darshan Karkar | Prof. Sweta Katariya

Area of research: Mechanical Engineering

Gearless Transmission Mechanisms, Often Referred To As Elbow Or Link-based Transmission Systems, Provide An Alternative Method Of Power Transfer Without The Use Of Conventional Gears. These Systems Utilize A Network Of Rotating Links And Joints To Transmit Motion And Torque From The Driving Shaft To The Driven Shaft. This Approach Reduces Mechanical Complexity, Minimizes Friction Losses Associated With Gear Teeth, And Lowers Maintenance Requirements. The Gearless Transmission Mechanism Is Particularly Useful In Applications Where Shafts Intersect At Various Angles And Smooth, Continuous Motion Is Required. This Study Discusses The Working Principle, Design Considerations, Components, Advantages, Limitations, And Potential Applications Of Gearless Transmission Systems. The Aim Is To Analyze Their Efficiency, Reliability, And Feasibility As A Substitute For Traditional Gear-based Transmissions In Mechanical And Industrial Systems.

Author: Surykar Pundlik Shivaji | Melkunde Manmath Shivsharan | Gade Prathamesh Keshavrao | Bhande Sakshi Ramdas | Prof. Rathod Yuvraj Shivaji

Area of research: Information Technology

Urban Waste Management Systems Face Challenges Such As Inefficient Route Planning, Delayed Monitoring, High Operational Costs, And Increased Carbon Emissions. EcoVision Is An Intelligent Waste Management Platform Designed To Optimize Urban Sanitation Workflows Using Cloud Infrastructure And Automation Technologies. The System Enables Real-time Data Access And Cross-regional Scalability, Ensuring Efficient Monitoring And Management Of Waste Collection Activities. The Core Innovation Lies In The Integration Of Robotic Process Automation (RPA), Which Automates Repetitive Administrative Tasks, Including Route Scheduling, Bin-level Monitoring Alerts, And Billing Processes, Without Human Intervention. By Reducing Manual.for, Optimizing Resource Utilization, EcoVision Minimizeesoperational Costs And Decreases Fuel Consumption Of Collection Vehicles, Thereby Reducing Carbon Emissions. The Proposed System Contributes To Building A Responsive, Scalable, And Sustainable Waste Management Ecosystem That Supports Smart City Initiatives And Cleaner Urban Environments.

Author: Abirami.N | Thanigaivelavan | Roshan N.S | Udhaya Saravanan

Area of research: Computer Science & Engineering

Author: K.Nansy | Jencia J | Chrisma Zion H

Area of research: Cyber Security

The Cyber Security Scenario Has Become A Matter Of Serious Concern In This Digital World, Where Organizations And Individuals Rely On Technology For Every Activity Ranging From Communication And Commerce To Data Storage. With The Ever-increasing Sophistication Of Cyber Threats, It Becomes Essential To Put In Place Strong Mechanisms For Defines. This Paper Focuses On The Future Of Cyber Security, Bringing Into View Emergent Attack Strategies And Protection Technologies. Social Engineering, Ransomware, And Advanced Persistent Threats (APTs) Are Among The Topics Of Interest. The Paper Also Discussed Proactive Measures In Security, Such As The Application Of Artificial Intelligence And Machine Learning In Improving Threat Detection And The Enhancement Of Response Times. It Also Examined The Part Which Ethical Hacking Plays In The Identification And Mitigation Of Vulnerabilities. Crowning All This Will Be The Affirmation Of The Need For An Integrated And Holistic Approach In Cyber Security; That Is, Combining The Technical With The User Aspect And Enforcing Policies Against Unavailable Threats. This Paper Is Intended To Contribute To The Continuing Defined Discourse On The Creation Of Secure Systems In An Increasingly Interconnected World.

Author: Rajkumar R | Vignesh G | Dharaneesh S R

Area of research: Library And Information Science

Resource Sharing Has Long Been A Fundamental Principle In Libraries And Information Centers, Allowing Institutions To Enhance Access To Knowledge While Reducing Costs. With The Rapid Progress Of New Information Technologies, Resource Sharing Has Moved Beyond Traditional Interlibrary Lending Methods To Dynamic, Networked, And Digital Environments. This Article Investigates The Concept Of Resource Sharing, The Role Of Developing Information Technologies In Reshaping Resource-sharing Practices, And The Difficulties And Opportunities That These Technologies Provide. It Investigates Digital Libraries, Cloud Computing, Networking Technologies, Consortiums, And Open-access Platforms As Important Drivers Of Modern Resource Sharing. The Study Indicates That New Information Technologies Improve Efficiency, Accessibility, And Collaboration, Making Resource Sharing More Sustainable And User-centered In The Digital Age.

Author: Balasaheb Kakasaheb Wayal

Area of research: Computer Science

Signature Verification Plays A Critical Role In Authentication Systems Used In Banking, Legal Documentation, And Financial Transactions. However, Skilled Signature Forgeries Pose A Significant Challenge For Traditional Verification Techniques. This Paper Presents A Deep Learning-based Approach For Detecting Skilled Signature Forgeries Using A Convolutional Neural Network (CNN). The Proposed System Compares An Original Signature With A Suspected Signature And Determines Whether The Signature Is Genuine Or Forged. The Model Is Implemented Using The PyTorch Deep Learning Framework And Deployed Through A Flask-based Web Application. Image Preprocessing Techniques Such As Resizing, Grayscale Conversion, And Normalization Are Applied Before Feeding The Signatures Into The CNN Model. The System Extracts Discriminative Features From Signature Images Through Multiple Convolutional Layers And Predicts The Authenticity Of The Signature With A Confidence Score. Experimental Results Demonstrate That The Proposed Approach Effectively Identifies Forged Signatures And Provides Reliable Verification Performance. The Developed System Can Assist In Preventing Fraud In Financial And Authentication Systems.

Author: Mrs. K. Menaka | Ms. S. Aarthi | Ms. K. Kaladevi | Ms. M. Kaviya

Area of research: BCA

Artificial Intelligence (AI) Has Become A Core Component In Modern Video Game Development, Significantly Enhancing The Behavior And Realism Of Non-Player Characters (NPCs). Traditional NPCs Relied On Scripted Rules And Predefined Logic, Resulting In Predictable And Repetitive Gameplay. Recent Advancements In Machine Learning, Reinforcement Learning, And Decision-making Algorithms Have Enabled NPCs To Exhibit Adaptive, Intelligent, And Human-like Behavior. This Paper Explores The Evolution Of NPC Intelligence, Key AI Techniques Used In Contemporary Games, And Real-world Case Studies Demonstrating Intelligent NPC Systems. The Study Also Analyzes The Advantages, Challenges, And Future Potential Of AI-driven NPCs In Creating Immersive And Dynamic Gaming Environments. The Findings Highlight How Intelligent NPCs Improve Player Engagement, Realism, And Replayability, Positioning AI As A Transformative Force In The Gaming Industry.

Author: Antony Cynthia | Nirmal P | Daniel Manovah Z | Mahinda Sivashanmugam

Area of research: Computer Science

This Paper Presents An Artificial Intelligence-based Automatic Expense Tracking System Designed To Help Users Manage Their Financial Activities Efficiently. Traditional Expense Tracking Methods Require Manual Input And Lack Predictive Insights About Future Spending. The Proposed System Automatically Collects And Categorises Expenses And Applies Machine Learning Algorithms To Predict Future Spending Patterns. The System Processes Historical Transaction Data And Identifies Spending Behaviour To Generate Meaningful Insights. A Dashboard Visualisation Using Charts And Graphs Allows Users To Analyse Their Expenses Effectively. Machine Learning Techniques Such As Linear Regression And Decision Tree Algorithms Are Used To Forecast Future Expenditures. Experimental Results Show That The Proposed Model Improves Financial Awareness And Helps Users Plan Their Budgets Effectively. The System Can Be Integrated With Mobile Applications And Banking APIs For Real-time Financial Monitoring. This Approach Enhances Personal Finance Management Through Intelligent Prediction And Automated Analysis.

Author: Harini R | S. Senthamaraiselvi

Area of research: Computer Science Engineering

Cloud Computing Has Become A Fundamental Platform For Storing Data And Running Applications Due To Its Scalability, Flexibility, And Cost Efficiency. However, The Rapid Growth Of Cloud Environments Has Also Increased Security Threats Such As Unauthorized Access, Malware Attacks, And Distributed Denial-of-service (DDoS) Attacks. Traditional Security Mechanisms Often Struggle To Detect New And Sophisticated Cyber Threats In Dynamic Cloud Infrastructures. To Address These Challenges, Intelligent Intrusion Detection Systems (IDS) Based On Machine Learning (ML) Techniques Have Gained Significant Attention. This Study Explores Various Machine Learning Approaches For Developing Intelligent Intrusion Detection Systems In Cloud Networks. Machine Learning Algorithms Such As Decision Trees, Support Vector Machines (SVM), Random Forest, Naïve Bayes, And Deep Learning Models Are Used To Analyze Network Traffic And Identify Abnormal Patterns That Indicate Potential Intrusions. These Models Are Trained On Network Datasets To Distinguish Between Normal And Malicious Activities With High Accuracy.

Author: Rendla Ramyakrishna | Veerender Aerranagula | Angoth Lakshman | Bhukya Vijay kumar

Area of research: Computer Science And Engineering

Road Accidents Are A Major Cause Of Injuries And Fatalities Worldwide, Often Resulting In Severe Consequences Due To Delays In Emergency Response And Medical Assistance. Immediate Detection And Notification Of Accidents Are Essential To Ensure Timely Intervention And Improve The Chances Of Survival For Victims. Traditional Accident Reporting Methods Depend On Manual Reporting By Witnesses Or Victims, Which May Not Always Be Possible In Critical Situations. This Project Proposes A Smart Accident Detection And Emergency Alert System Using Mobile Sensors That Utilizes The Built-in Sensors Available In Modern Smartphones To Automatically Detect Accidents And Notify Emergency Contacts. The System Monitors Sensor Data Such As Accelerometer, Gyroscope, And GPS Location To Identify Sudden Impacts, Abnormal Motion Patterns, Or Abrupt Changes In Movement That May Indicate A Potential Accident. When Such An Event Is Detected, The System Automatically Triggers An Alert Mechanism That Sends An Emergency Message Along With The Victim’s Real-time Location To Predefined Contacts Or Emergency Services. By Integrating Sensor Data Analysis With Real-time Location Tracking, The Proposed System Provides A Reliable And Cost-effective Solution For Automatic Accident Detection And Emergency Alert Generation. The System Aims To Reduce Emergency Response Time, Improve Road Safety, And Enhance The Chances Of Timely Medical Assistance For Accident Victims.

Author: Mr. Mariya John | Mr.A. Aravinth | Mr.A. Haripragash | Mr.L.S. Dharaneesh

Area of research: Computer Science And Engineering

Atherosclerosis Is A Chronic Cardiovascular Condition Characterized By The Gradual Accumulation Of Plaque Within Arterial Walls, Resulting In Restricted Blood Flow And Increased Risk Of Heart Attacks And Strokes. Early Detection Is Essential For Effective Clinical Intervention; However, Traditional Diagnostic Techniques Such As Coronary Angiography, CT Scans, MRI, And Ultrasound Rely Heavily On Manual Interpretation. These Methods Are Labor-intensive, Prone To Inter-observer Variability, And Often Detect The Disease At Advanced Stages. This Project Proposes An AI-based Atherosclerosis Detection Framework Leveraging Machine Learning (ML) And Deep Learning (DL) Techniques To Automatically Analyze Cardiovascular Imaging Data, Identify Plaque Regions, Classify Disease Severity, And Support Clinical Decision-making. The System Integrates Image Preprocessing, Feature Extraction, And Convolutional Neural Network (CNN)-based Analysis To Detect Early-stage Plaques With High Accuracy. Real-time Implementation Allows Continuous Monitoring And Early Alerts, Reducing Cardiovascular Risk. Testing On Annotated Datasets Demonstrates Improved Diagnostic Performance, Including Higher Sensitivity, Precision, And Reduced False Positives, Compared To Conventional Methods.

Author: Mr.S. Balaji3= | Mr.S. Ganeshkumar | Mr.V. Deepak | Mr.S. Senthilkumar

Area of research: Deep Learning

Lung Fibrosis Is A Chronic And Progressive Respiratory Disease Characterized By Scarring Of Lung Tissue, Leading To Reduced Lung Function And, In Severe Cases, Respiratory Failure. Early Detection Is Critical For Timely Intervention And Improved Patient Outcomes. Traditional Diagnostic Methods Rely Heavily On Manual Interpretation Of Imaging Data And Clinical Records, Often Causing Delays Due To Subtle Early-stage Manifestations. This Project Proposes An AI-enhanced Early Lung Fibrosis Detection Platform That Integrates Medical Imaging And Patient Clinical Records To Improve Diagnostic Accuracy And Efficiency. Using A Multimodal Data Fusion Approach, The System Combines Imaging Features With Electronic Health Record Insights To Classify Patients Into Categories Such As Normal, Early-stage Lung Fibrosis, And Advanced Fibrosis. By Providing Intelligent Decision Support, The Platform Assists Clinicians In Early Diagnosis, Reduces Ambiguity, And Promotes Faster Treatment Planning.

Author: Mrs. Kavitha Pandian | Mr.J. Rajesh | Mr.M.PrithiviRaj | Mr.M. Sakthivel

Area of research: Artificial Intelligence

Artificial Intelligence (AI) Has Emerged As A Crucial Element In Contemporary Cyber Security Frameworks, Facilitating Automated Threat Detection, Predictive Analytics, And Adaptive Defense Strategies. Nonetheless, The Deployment Of AI-driven Systems Introduces Novel Categories Of Vulnerabilities That Surpass Conventional Technical Attack Surfaces. A Significant Emerging Concept Is The Psychological Attack Surface, Which Encompasses Cognitive, Emotional, And Behavioral Vulnerabilities That Can Be Exploited By Attackers Via Digital Communication Platforms And AI-enabled Systems. Traditional Cyber Security Paradigms Primarily Emphasize Software Vulnerabilities, Network Breaches, And System Misconfigurations, Frequently Overlooking The Psychological Dimensions That Affect Human Decision-making In Cyber Contexts. Recent Research Indicates That A Considerable Fraction Of Cyber Attacks Depend On Social Engineering Methods That Manipulate Trust And Emotional Responses, Rather Than Relying On Technical Deficiencies [1], [10]. The Widespread Implementation Of AI Technologies, Such As Conversational Agents, Recommendation Systems, And Emotion Recognition Frameworks, Enables Adversaries To Take Advantage Of Both Human Users And AI Systems By Utilizing Psychological Triggers Such As Urgency, Fear, Authority, And Social Trust [2], [3]. Moreover, Adversarial Attacks Aimed At Machine Learning Models Have Broadened The Attack Surface Related To Intelligent Systems [4], [5]. These Advancements Underscore The Necessity For Cyber Security Frameworks That Incorporate Behavioral Intelligence And Psychological Assessment. This Paper Introduces An Innovative Framework Referred To As The Psychological Attack Surface Model (PASM), Which Merges Artificial Intelligence Methodologies With Principles Of Cyber Psychology To Identify And Mitigate Psychologically Driven Cyber Threats. The Proposed Model Encompasses Layered Components, Including Psychological Data Collection, Behavioral Signal Analysis, Attack Surface Mapping, And AI-driven Defense Strategies. Machine Learning Algorithms And Natural Language Processing Methods Are Employed To Recognize Manipulation Patterns Within Communication Settings. This Framework Facilitates Dynamic Modeling Of Psychological Vulnerabilities In Human–AI Interaction Ecosystems And Supports Proactive Identification Of Adversarial Behaviors. By Amalgamating Insights From Artificial Intelligence Security And Cyber Psychology Studies [6], [9], This Research Contributes To The Establishment Of Human-centered Cyber Security Architectures Capable Of Addressing Emerging Socio-technical Risks.

Author: Karthick kumar A | Aaron lee peter | Dharanesh R L | Raj Kumar R

Area of research: Mechanical Engineering

The Project Applied The Use Of Observation Based On The Manual Method Currently Used Using Poisoning Of Various Pests. The Objective Of This Project Is To Design A Device That Is Capable Of Producing A More Effective Pesticide Sprayer For Use In Small Or Rural Industries In The Agricultural Sector. Additionally, There Are Several Research Scopes That Have Been Defined In This Project, Producing And Developing Ergonomic Wheel Sprayers. To Reduce Spraying Time In Vegetable Gardens Or Orchards And To Increase Spraying Efficiency As It Contains More Than One Nozzle During Spraying. All These Are Set To Solve Some Of The Problems That Arise With The Use Of Existing Methods Among Which, The Existing Sprays Cannot Be Effective And Require Additional Time For Spraying. The Material For This Project Also Requires Special Properties That Do Not Rust And Do Not Affect Plants, Based On The Literature Review Conducted Stainless Steel Is The Most Suitable For This Project. While For The Component Formation Process, The Research Methodology Is Used For The Project Production Process By Using Flow Charts As A Guide To Plan The Production And Testing Of The Project. As A Result, The Whole Project Was Successfully Produced With The Additional Rate Of Time Saving Of Traditional Methods. Based On These Results, The Results Of Analysis And Discussions Conducted, It Can Be Concluded That This Sprayer Wheel Has Achieved The Objectives Discussed. In Addition, This Tool Is Also Proven To Be Able To Save Time Differently The Traditional Way.

Author: Pande Apurv Virendra | Patil Sachin Shankar | Surwase Ram Vijaykumar | Hake Anuradha Anandrao | Prof. Sabde Abhijit Manoharrao

Area of research: Mechanical Engineering

This Project, Titled “Manual Pedal Press Pneumatic Lifter Jack,” Involves The Design And Fabrication Of A Manually Operated Pneumatic Lifting System With A Lifting Capacity Of 120–145 Kg. The System Uses A 191 Psi Foot Pump, A Direction Control Valve, And A Double-acting Cylinder To Lift And Lower Loads Efficiently. The Working Principle Is Based On Compressed Air Supplied Through A Pedal-operated Pump, Which Is Directed By The Control Valve Into The Cylinder To Produce Linear Motion. The Structure Is Fabricated Using Mild Steel To Provide Sufficient Strength And Stability During Operation. The Main Objective Of This Project Is To Develop A Cost-effective, Portable, And Easy-to-operate Lifting Device Suitable For Workshops, Garages, And Small Industries. The Pneumatic Lifter Jack Reduces Manual Effort, Improves Safety, And Ensures Smooth Lifting Compared To Conventional Mechanical Jacks. Testing And Observation Confirm That The System Performs Reliably Within The Specified Load Range, Demonstrating The Practical Application Of Pneumatic Technology In Light Material Handling Operations.

Author: Rajure Sanket Rajendra | Dhole Pushkar Shripad | Patil Omkar Eknath | Shete Shivshankar Guptling | Prof.Sabde Abhijit Manoharrao

Area of research: FINANCE

Capital Structure Plays A Vital Role In Determining A Company’s Financial Stability And Long-term Profitability. The Mix Of Debt And Equity Used To Finance Business Operations Directly Influences Risk, Return, And Shareholder Value. This Study Examines The Impact Of Capital Structure On The Profitability Of ITC Ltd., One Of India’s Leading Diversified Conglomerates With A Strong Presence In FMCG, Hotels, Paperboards, Packaging, And Agri-business. The Primary Objective Of This Research Is To Analyse How Variations In Debt-equity Composition Affect Key Profitability Indicators Such As Return On Equity (ROE), Return On Assets (ROA), Net Profit Margin, And Earnings Per Share (EPS). The Study Is Based On Secondary Data Collected From The Annual Reports Of ITC Ltd., Financial Statements, And Relevant Financial Databases Over A Selected Period. The Findings Indicate That ITC Ltd. Has Traditionally Maintained A Conservative Capital Structure With Minimal Reliance On External Debt. This Low Leverage Strategy Has Contributed To Financial Stability And Consistent Profitability, While Reducing Financial Risk. However, The Study Also Explores Whether Optimal Utilization Of Debt Could Potentially Enhance Shareholder Returns Without Significantly Increasing Financial Distress. The Case Of ITC Ltd. Highlights How Strategic Financial Management Supports Long-term Value Creation While Preserving Financial Flexibility.

Author: S. Alahappan | Dr. P. Pirakatheeswari

Area of research: AI & Machine Learning

Retrieval-Augmented Generation, Large Language Models, AI Chatbot, Information Retrieval, Natural Language Processing.

Author: S Thilaiyarasi | RakeshK | BharathiE | Stephen Karthikeyan K

Area of research: Cybersecurity And Network Security

Network Security Has Become A Critical Concern As Modern Organizations Rely Heavily On Interconnected Systems And Digital Infrastructure. Identifying Vulnerabilities In Network Services Is Essential To Prevent Unauthorized Access, Data Breaches, And System Compromise. Traditional Vulnerability Assessment Processes Often Require Manual Analysis Using Multiple Tools, Which Can Be Time-consuming And Prone To Human Error.This Paper Presents Cyber-Net-SecX, An Automated Framework Designed To Streamline The Process Of Network Vulnerability Assessment. The Proposed System Integrates Network Reconnaissance, Vulnerability Intelligence Extraction, Exploit Feasibility Validation, And Risk Analysis Into A Single Automated Workflow. The Framework Utilizes Nmap For Host Discovery, Port Scanning, And Service Enumeration, While Detected Services Are Correlated With Common Vulnerabilities And Exposures (CVE)entries To Identify Potential Security Weaknesses. To Determine Exploit Feasibility, The System Connects To TheMetasploit Framework UsingRemote Procedure Call (RPC) And Verifies The Availability Of Relevant Exploit Modules. A CVSS-based Risk Analysis Model Is Applied To Evaluate Vulnerability Severity And Classify Risks.The Implementation Generates A Structured Multi-page Security Assessment Report Containing Detected Vulnerabilities, Exploit Intelligence, And Risk Severity Visualization. Experimental Testing In A Controlled Environment Using A Vulnerable Virtual Machine Demonstrates That The Automated Framework Significantly Reduces Manual Analysis Time While Improving Vulnerability Detection Efficiency. The Results Highlight The Effectiveness Of Automated Security Assessment Frameworks In Supporting Modern Cybersecurity Operations.

Author: Harish R | Harish S | Hariharan C | Eshwara K

Area of research: Civil Engineering

The Construction Sector Creates A Significant Amount Of Jobs And Is Essential To The Nation's Socioeconomic Development.The Aim Of This Research Is To Determine And Examine The Key Elements That Impact Construction Project Performance By Causing Time And Expense Overruns. Numerous Aspects Are Covered In This Study, Including Inadequate Project Planning And Scheduling, Issues With Subcontractors, Poor Site Management And Supervision, Material Management Issues, A Lack Of Collaboration Among Stakeholders, Etc. According To The Study's Conclusions, The Ishikawa Diagram Is A Valuable Tool For Determining And Analysing The Causes And Effects Of Labour, Material, And Equipment-related Delays. Thus, It Can Assist Project Managers In Ensuring That The Project Is Completed Smoothly And Within The Budget And Time Frame That Have Been Set.Implementing Efficient Material Management Is Essential For Timely Procurement And Inventory Issuance To Minimise Delays Caused By Material Shortages, As Materials Account For Roughly 70% Of The Total Cost Of Construction. Its Utilisation Can Be Maximised With No Waste By Implementing Resource Levelling And Smoothing. According To This Study, Building Projects May Be Made To Be Completed On Time And Within Budget By Using Contemporary Project Management Tools Like Microsoft Project, Primavera, Newton Software, And Others To Effectively Monitor The Project Schedule.

Author: Shiv Kumar Sahu | Bikesh Tripathi

Area of research: Civil Engineering

The Environment And Transportation Are Closely Related And Reliant On Each Other. In Addition To Promoting Economic Growth, Sustainable Transportation Initiatives—like Cleaner Urban Transportation Systems And More Effective Rural Road Rehabilitation—also Have Significant Social Advantages. However, The Environment And Nearby Communities May Be Significantly Impacted By Transport Projects. The Design Of Roadway Alignments Is Typically Predicated On Minimising Costs, Including Earthwork Costs (cutting And Filling). Road Accidents And Highway Longevity Are Two Examples Of The Issues That Are Taken Into Consideration When Designing A Decent And Sustainable Route Alignment. These Elements Influence The Creation Of A Sustainable Road Alignment. Roads Built In A Nation's Mountainous Areas Are Referred To As Ghat Roads Or Hillroads. In Terms Of Alignment, Design, Building, And Maintenance, These Roads Pose Significant Challenges. Hill Roads Are More Prone To Accidents Due To Their Curves, Sharp Turns, Steep Grades, And Narrow Roadway Width. Furthermore, A Hill Road's Construction And Upkeep Are Severely Impacted By Prolonged Rain. Heavy Rains Can Cause Landslides And Slips At Numerous Spots Along The Hill Routes. Therefore, In Order To Build A Sturdy And Safe Road, Much Attention Must Be Taken Throughout Its Layout And Construction. Moreover, A Large Number Of Streams Cross The Road, And Hence A Suitable Facility For Cross Drainage Is Needed. In This Project, We Are Designing A Hill Road Gauharganj Stretch Through A Distance Of 6.1 Km

Author: Natasha Meshram | Hitesh Kodwani

Area of research: Commerce With Professional Accounting

The Rapid Growth Of Digital Technology And E-commerce Platforms Has Significantly Changed The Buying Behavior Of Consumers. At The Same Time, Traditional Offline Shopping Continues To Remain Important Due To Personal Interaction, Product Inspection, And Immediate Purchase. This Study Aims To Compare Online And Offline Buying Behavior Of Consumers And Identify The Key Factors Influencing Their Purchase Decisions. The Research Focuses On Aspects Such As Convenience, Price, Trust, Product Availability, Customer Satisfaction, And Perceived Risk In Both Modes Of Shopping. Primary Data Were Collected Through A Structured Questionnaire From Selected Respondents, And Secondary Data Were Gathered From Journals, Books, And Websites. Statistical Tools Such As Percentage Analysis And Comparative Analysis Were Used To Interpret The Data. The Findings Of The Study Highlight Significant Differences In Consumer Preferences Between Online And Offline Shopping. The Study Provides Useful Insights For Marketers And Retailers To Improve Their Strategies By Understanding Consumer Expectations And Integrating Both Online And Offline Channels Effectively.

Author: Dr W. Saranya | Mr. S.Venkatesh

Area of research: Computer Science And Engineering

Vehicle Breakdown Is A Common Problem Faced By Vehicle Users During Travel. When A Vehicle Breaks Down Unexpectedly, It Becomes Difficult For Users To Find Nearby Mechanics Quickly. This Causes Delay, Inconvenience, And Safety Risks. The Smart Road Assist System Is Developed To Solve This Problem By Providing Real-time Assistance To Vehicle Users. The System Detects The User's Current Location Through The Browser And Displays Nearby Available Mechanics. Users Can Send Service Requests, And Mechanics Can Accept The Request Through The System. The System Also Provides Live Tracking, Secure Login Using JWT Authentication, And Real-time Communication. This Solution Improves Response Time, Reduces User Stress, And Makes Vehicle Assistance More Efficient And Reliable.

Author: Mrs. N. Sathiyarani | Ms.C.Mareeswari | Ms.N.Sevanthihaa | Ms. D.Seethalakshmi

Area of research: Civil Engineering

The Efficient Use Of Scheduling Tools, Such As Gantt Charts, To Plan And Then Provide Progress Reports Within The Project Context Is Correlated With Project Planning, Which Is A Crucial Component Of Project Management. The Numerous Challenges Encountered Throughout The Project Were Also Recognised And Resolved. Additionally, The Conventional Method Is Time-consuming. With The Aid Of A Case Study Of A Project Being Carried Out In India, The Given Work Will Provide Them The Chance To See The Benefits Of Microsoft Project (MSP), Which Expedites Construction And Makes The Project Cost-effective With Proper Planning. An Examination Of A High-rise Residential Building Located In Bicholi Mardana, Indore (M.P.), Was Taken Into Consideration For The Current Project. Additionally, Utilising The Blueprints And Drawings Provided By The Site Officials, A Detailed Estimate Was Supplied. The Project's Overall Length Was Determined, And Microsoft Project Software Was Used To Schedule The Tasks And Activities.

Author: Komal Malviya | Bikesh Tripathi

Area of research: Civil Engineering

The Construction Sector Is Increasingly Challenged To Approximately 10% Of All Road Fatalities Worldwide Occur In India, Where 1,46,133 People Are Killed Annually. The Identification Of Blind Spots In Indore And Other Current And Possible Safety Risks To Drivers Are The Main Topics Of The Case Study. Like The Majority Of India's 3.3 Million Km Of Existing Road Networks, The Section Was Selected For The Road Safety Audit (RSA) Study Because It Already Existed And Sees A Significant Volume Of Traffic Year-round.Geometry And Alignment Adjustments Of Any Type Are Difficult To Implement And Could Otherwise Be Highly Costly. In Order To Make The Travel Safer For Road Users And The Surrounding Region, This Audit Identified Possible Road Safety Issues And Offered Recommendations To Stop Frequent Accidents Or At Least Lessen Their Severity. For Authorities And Road Users, The Fact That India Has 11% Of The World's Automobiles And 11% Of Traffic Fatalities Is Extremely Alarming. By Identifying Black Spots, We Can Implement Corrective Safety Measures In Regions That Are Prone To Accidents. In Order To Analyse The Severity Of Accidents, We Conducted A Road Safety Audit (RSA) After Identifying Areas That Are Prone To Accidents. An Official Evaluation Of New Or Existing Roads And Areas Adjacent To Roads From The Perspective Of All Road Users Is Called A Road Safety Audit (RSA), And Its Objective Is To Find Potential Crash Sites And Safety Deficiencies. In This Paper Presenting Review Of Literatures

Author: Bipin Kumar Gupta | Praveen Ghdode

Area of research: Civil Engineering

In Indian Cities, The Need For Transport Has Increased As A Result Of Rapid Urbanisation. When It Comes To Meeting The Increasing Travel Demand, Public Transport Has Fallen Short. Population Profiling Is Increasing The Number Of People Who Own Cars. On Indian City Highways, Congestion Is Caused By A Significant Number Of Private Vehicles, A Variety Of Traffic Patterns, And A Shortage Of Available Road Space. The Metropolis Of Bhopal Is Expanding Quickly. Bhopal Roads See Considerable Traffic Due To A Combination Of A High Demand For Transit And A Shortage Of Available Road Space. The Majority Of The Population's Access To Transit Services Is Becoming More Difficult Due To Bhopal City's Explosive Growth. The Expanding Traffic Demand Cannot Be Accommodated By The Existing Infrastructure. The Purpose Of This Study Is To Examine How Various Factors That Lead To Traffic Jams In Cities Interact With One Another. Due To Traffic Congestion, A Large Portion Of Working Hours Is Wasted On The Roads, Which Negatively Affects The Economy As A Whole. Numerous Studies And Works Of Literature Have Been Devoted To The Examination Of Congestion And Its Consequences. But The Final Result Hasn't Been Good Enough. The Goal Of The Current Study's Congestion Projection Is To Identify The Underlying Viability Of The Diversified Traffic Situation And Offer Better Guidance For Controlling And Preventing These Protracted Traffic Jams In Mixed Traffic With No Lane Discipline.

Author: Dharmendra Kumar Sahu | Praveen Ghdode

Area of research: Civil Engineering

When There Is A Greater Demand For Space Than The Road Can Provide Due To An Inflow Of Vehicles, Traffic Jams Arise. Statistical Computer Models And Numerical Field Data About Vehicle Counts Are Essential Instruments For Assessing The Traffic Volume And Serviceability Of A Route. In This Investigation, Both Have Been Employed. This Study's Main Goals Are To Assess The Level Of Service (LOS), Travel Time, Degree Of Saturation, Wait Time, And Perform A Thorough Analysis Of The Traffic Flow To Develop A Plan To Lessen Roundabout Traffic Congestion. During Peak Hours, From 12:00 To 1:30 P.m., The Number Of Vehicles Passing Through The Roundabout At Redisson Square's Eastern Entrance, Located At MR 10, Was Counted. In Order To Make The Mixed Traffic Stream Uniformly Equal, It Was Subsequently Changed To Passenger Car Units (PCU). The Roundabout At Redisson Square At The City Entry Needs More Development To Improve The Terrible Traffic Conditions, According To The Findings Of The Systematic Study Of The Data Collected. The Current Road Layout Was Measured Using Surveying Equipment To Perform An Accurate Simulation, And The Findings Were Utilised To Create A Model. Lanes For Acceleration And Deceleration Were Suggested To Enhance Roundabout Efficiency And Ease Traffic. Following Road Safety, The Roundabout Extension Idea Was Then Incorporated Into The Geometric Design In Order To Assess Its Efficacy In Easing Traffic Congestion. From 2015 To 2024, The Roundabout's Lanes' Level Of Service Improved Significantly In Terms Of Traffic Flow, Road Safety, And Accident Records.

Author: Suraj Bhagoriya | Hitesh Kodwani

Area of research: Civil Engineering

The Two Primary Forms Of Pavement Are Rigid Pavement And Flexible Pavement. Bituminous Is Used As The Aggregate And Binder Material In Flexible Pavements. The Gelatinous, Viscous Mixture Of Hydrocarbons Known As Bitumen, Which Is Used For Roofing And Pavement Materialisation, Can Be Obtained Naturally Or As A Byproduct Of Petroleum Refinement. Globally, Bitumen Is Utilised As A Binder For Flexible Pavements. When Heated, Bitumen Turns Poisonous And Has Negative Effects On The Environment, Even Though It Is Not Dangerous In Normal Circumstances. Additionally, The Production Of Bitumen, Which Comes From Petroleum, A Non-renewable Energy Source, Will Have Caused The Depletion Of Petroleum Reserves. The Highway Sector Faces A Significant Challenge In Reducing Its Reliance On Fossil Fuels And Recycling Its Trash. With Bitumen, A Binder Made From Petroleum, Being One Of Its Primary Constituents, The Asphalt Industry Is Certainly One With A Sustainable Environmental Impact. The Production Of Bitumen Contributes To Massive Emissions Of Carbon Dioxide, Which Harm The Environment. Utilising Waste Oils As Substitute Binders Is The Subject Of This Research Project. Waste Cooking Oil And Waste Engine Oil Are The Waste Oils Used. These Are Investigated And Evaluated To Create A Sustainable Environment. This Project Will Offer A Substitute Or Modified Binder And A More Effective Method For The Secure Disposal Of Waste Oils Produced. Therefore, This Initiative Is Advantageous In Terms Of The Safe Disposal Of Waste Oils As Well As The Environmental Elements Of The Alternative Binder.

Author: Manoj Ahirwar | Hitesh Kodwani

Area of research: Computer Science

Human–wildlife Conflict Is A Growing Global Concern As Expanding Human Settlements Encroach Upon Natural Habitats. Traditional Wildlife Monitoring Methods—manual Patrols And Passive Camera Traps—suffer From Delayed Response, Limited Scalability, And An Inability To Assess Threat Severity In Real Time. This Paper Presents An Intelligent Wild Animal Detection And Adaptive Risk Alert System That Integrates YOLOv8-based Real-time Object Detection, DeepSORT Multi-object Tracking With Persistent Unique Identification, Trajectory-based Behavior Classification (roaming, Approaching, Aggressive), A Novel Five-factor Composite Risk Scoring Engine With Sigmoid Normalization, Species-aware Dynamic Geofencing, Automated Sound Deterrents, And Multichannel Alert Dispatch Via SMS And WhatsApp. The System Is Orchestrated Through A Modular Python Pipeline And Visualized On A Live FastAPI Web Dashboard With WebSocketdriven Video Streaming. Experimental Evaluation On Wildlife Surveillance Footage Demonstrates A Mean Average Precision (mAP@0.5) Of 87.3% For Detection, 94.6% Tracking Consistency (MOTA), And Sub-200ms End-to-end Latency Per Frame On Consumer-grade Hardware. The Adaptive Risk Formula— Rfinal= σ α•Sd•ψ(b)+β•(1−dnorm)•v+γ•log2(n+1)+ —enables Context-sensitive Threat Es- Calation, Reducing False-positive Alerts By 41% Compared To Static Threshold Baselines. The Platform’s Modular Architecture And YAML-driven Configuration Support Rapid Deployment At Wildlife Corridors, Forest Perimeters, And Agricultural Buffer Zones.

Author: Rajaram K | Juliet S | Ramya M | Anish I

Area of research: Artificial Intelligence And Data Science

The Rapid Growth Of The Technology Sector Has Intensified The Demand For Effective Interview Preparation Tools That Can Help Candidates Develop Both Technical Knowledge And Soft Skills. Traditional Interview Preparation Methods, Including Coaching Sessions And Static Question Banks, Often Lack Personalized Feedback And Real-time Performance Evaluation. This Paper Presents An AI Interview Practice Partner, An Intelligent Web-based Mock Interview System That Leverages Natural Language Processing (NLP) And Generative AI To Simulate Realistic Interview Environments. The Proposed System Enables Users To Select Interview Categories Suchas Technical, HR, Or Aptitude-based Sessions At Varying Difficulty Levels. User Responses Are Analyzed In Real Time Using NLP Techniques For Relevance, Clarity, Correctness, And Logical Flow. Generative AI Models Power Dynamic Question Generation And Conversational Interview Flow. The System Provides Instant Personalized Feedback, Performance Scoring Across Multiple Dimensions Including Communication, Technical Knowledge, Confidence, And Behavioral Attributes. Experimental Results Demonstrate That The Proposed System Significantly Enhances User Interview Readiness, Reduces Anxiety, And Improves Performance Consistency Compared To Existing Static Platforms. The Architecture Follows A Client-server Model With Modular Components For Authentication, Question Generation, Response Evaluation, And Feedback Delivery, Ensuring Scalability And Extensibility For Future Enhancements Such As Voice-based Interviews And Emotion Detection.

Author: Gugan J | Gowtham M | Akilesh R | Varunkumar S

Area of research: Logistics

The Construction Sector Is Increasingly Challenged To Minimize Environmental Harm, Preserve Natural Resources, And Address The Growing Accumulation Of Industrial And Construction Waste. Sustainable Building Structure Using Industrial Waste, Recycling And Reuse Of Construction Material Examines The Potential Of Incorporating Materials Like Fly Ash, Slag, Silica Fume, And Construction And Demolition Waste As Eco-friendly Replacements For Traditional Building Materials. The Study Focuses On Evaluating Their Physical Characteristics, Structural Behavior, And Long-term Durability To Determine Their Effectiveness In Practical Construction Applications. Also, By Encouraging Recycling And Reuse, This Project Seeks To Reliance On Natural Aggregates, And Advance Circular Economy Principles Within The Industry. It Also Analyses The Economic Advantages, Including Lower Material Costs, Reduced Landfill Disposal, And Improved Waste Management Practices. The Results Indicate That Recycled Waste Materials Can Achieve Reliable Structural Performance While Offering Notable Environmental And Financial Benefits. Overall, The Study Shows That The Use Of Industrial And Construction Waste In Building Systems Is A Viable Strategy For Promoting Sustainable Infrastructure And Ensuring More Responsible Utilization Of Available Resources. Experimental Testing Was Carried Out To Evaluate Compressive Strength, Durability, And Material Performance Of Waste-incorporated Concrete Mixes.

Author: Lalithambiga K | Dr P Syam Sundar

Area of research: MBA

Warehouse Management And Material Handling Are Critical Functions That Influence Operational Efficiency In Manufacturing Organizations. The Purpose Of This Study Is To Examine Warehouse Management Practices And The Effectiveness Of Material Handling Systems Within Manufacturing Environments. The Research Focuses On Key Aspects Such As Inventory Control Procedures, Warehouse Layout Planning, Storage Methods, And The Use Of Handling Equipment And Digital Systems. The Study Identifies Operational Challenges While Also Exploring Opportunities For Improving Warehouse Efficiency. The Findings Suggest That Well-structured Warehouse Management Systems And The Adoption Of Modern Material Handling Techniques Significantly Enhance Accuracy, Reduce Operational Costs, And Improve Workflow Efficiency. The Study Also Offers Practical Suggestions That Can Help Organizations Strengthen Warehouse Operations And Improve Overall Supply Chain Performance.

Author: Nanthithaa M | Dr. P Syamsundar

Area of research: Human Resources

Employee Job Satisfaction Plays A Vital Role In Determining Organizational Success, Employee Retention, And Overall Productivity. In The Modern Business Environment, Organizations Are Increasingly Recognizing That Satisfied Employees Are More Committed, Motivated, And Willing To Contribute Beyond Their Assigned Responsibilities. This Study Aims To Analyse The Key Factors Influencing Employee Job Satisfaction, With Particular Emphasis On The Impact Of The Work Environment. Primary Data Was Collected From Employees Using A Structured Questionnaire, And Appropriate Statistical Tools Such As Descriptive Analysis And Regression Analysis Were Applied To Interpret The Findings. The Reliability Of The Instrument Was Tested Using Cronbach’s Alpha, Which Confirmed Good Internal Consistency Of The Measurement Scale. The Results Of The Study Reveal A Strong And Statistically Significant Positive Relationship Between Work Environment And Job Satisfaction. The Findings Indicate That Improvements In Workplace Conditions, Leadership Support, Communication Practices, And Overall Organizational Culture Significantly Enhance Employee Satisfaction Levels. The Study Highlights The Importance Of Creating A Supportive And Healthy Work Environment As A Strategic Priority For Management. By Focusing On Employee Well-being And Engagement, Organizations Can Strengthen Morale, Reduce Turnover Intentions, And Improve Long-term Organizational Performance.

Author: Don Damian P.D | Dr.P.Syamsundar

Area of research: Artificial Intelligence

The Prevailing Trajectory Of Artificial Intelligence Development Has Largely Equated Progress With Scale, Prioritizing Increasingly Larger Models Trained On Vast Datasets. While This Paradigm Has Delivered Notable Performance Gains, It Has Also Exposed Fundamental Limitations In Real-world Deployment, Particularly In Environments Constrained By Energy, Latency, Infrastructure, And Regulatory Boundaries. This Paper Argues For A Paradigm Shift Toward Adaptive AI Systems That Dynamically Align Computational Effort With Contextual Demands. Rather Than Maximizing Intelligence Uniformly, Adaptive AI Emphasizes Situational Adequacy—modulating Inference Depth, Resource Usage, And Decision Complexity In Real Time. The Study Examines How Such Systems Can Be Architected To Operate Efficiently Across Edge Devices, Mobile Platforms, And Distributed Industrial Settings Without Sacrificing Reliability Or Accountability. By Analyzing Current Deployment Constraints And Emerging Adaptive Design Principles, This Work Highlights How Intelligence Can Be Delivered Where And When It Is Needed, Rather Than Where Computation Is Cheapest. The Paper Positions Adaptive AI As A Critical Foundation For Sustainable, Responsible, And Scalable Intelligence, Capable Of Bridging The Gap Between Laboratory Innovation And Operational Reality[5].

Author: Daniel Manovah Z | Mahinda Sivashanmuga. | Nirmal P

Area of research: B.com Business Process Services

In The Past Decade, The Rapid Growth Of The Internet And Digital Technologies Has Transformed The Way People Shop. Traditional Retail Methods, Where Customers Physically Visit Stores, Are Increasingly Being Replaced By Online Shopping Platforms. E-commerce Platforms Like Amazon, Flip-kart, Anibal, And Mantra Provide Consumers With Access To Thousands Of Products Across Various Categories, From Electronics To Clothing, In Just A Few Clicks. While This Growth Has Made Shopping Convenient, It Has Also Introduced New Challenges. Customers Are Often Faced With An Overwhelming Number Of Choices, Making It Difficult To Select Products That Best Suit Their Needs. In Such A Competitive Environment, Businesses Are Constantly Searching For Innovative Methods To Attract And Retain Customers. One Of The Most Significant Solutions In Recent Years Has Been The Use Of Artificial Intelligence (AI) In Online Shopping. Artificial Intelligence, Broadly Defined, Refers To Computer Systems Capable Of Performing Tasks That Typically Require Human Intelligence.

Author: Dr. S. NAMASIVAYAM | K ASWINI

Area of research: MARKETING

The Rapid Growth Of Social Media Usage Has Significantly Transformed Marketing Practices, Particularly In The Skincare Industry. Platforms Such As Instagram, YouTube, And Facebook Have Become Powerful Channels For Product Promotion, Consumer Engagement, And Brand Communication. Among Various Digital Strategies, Influencer Marketing Has Emerged As A Highly Effective Approach For Promoting Online Skincare Products And Shaping Consumer Buying Behaviour. Influencers Create Product Awareness, Demonstrate Usage, Share Personal Experiences, And Provide Recommendations That Consumers Often Perceive As Trustworthy And Relatable. This Study Aims To Examine The Impact Of Influencer Marketing On Consumers’ Purchase Decisions Toward Online Skincare Products. It Analyses Key Influencing Factors Such As Influencer Credibility, Expertise, Attractiveness, Authenticity, Trustworthiness, And Electronic Word-of-mouth (e-WOM). In Addition, The Study Explores How Consumers Evaluate Other Aspects—including Product Ingredients, Price, Brand Reputation, Customer Reviews, And Perceived Risk—before Making A Purchase Decision. The Study Also Finds That Consumers Are More Likely To Be Influenced By Relatable Influencers Who Provide Honest Reviews And Demonstrate Genuine Product Usage. Furthermore, Transparency In Sponsored Content And Alignment Between Influencer Image And Product Type Strengthen Consumer Confidence. The Study Concludes That Influencer Marketing Plays A Crucial Role In Shaping Consumer Buying Behaviour In The Online Skincare Market. Credible, Authentic, And Knowledgeable Influencers Can Effectively Drive Consumer Engagement And Conversion, Making Them Valuable Partners For Skincare Brands In The Competitive Digital Marketplace.

Author: Sanjay.M | Dr.P.Pirakatheeswari

Area of research: B. Com Business Process Services

In Today’s Highly Competitive And Dynamic Marketplace, Advertising Plays A Crucial Role In Influencing Consumer Awareness, Attitudes, And Purchase Decisions. With The Rapid Expansion Of Markets And Increasing Diversity In Consumer Profiles, Businesses Are Constantly Seeking Innovative Ways To Communicate Effectively With Their Target Audiences. One Such Powerful And Increasingly Significant Approach Is Regional Language Advertising, Which Focuses On Delivering Marketing Messages In The Local Language Of Consumers. This Strategy Has Gained Substantial Importance, Particularly In Countries Like India, Where Linguistic Diversity Is Deeply Embedded In Cultural Identity And Daily Communication. India Is A Multilingual Nation With Over 22 Officially Recognized Languages And Hundreds Of Regional Dialects. While English And Hindi Have Traditionally Dominated Mass Media Advertising, A Growing Number Of Brands Are Now Embracing Regional Languages Such As Tamil, Telugu, Malayalam, Kannada, Marathi, Bengali, And Others To Connect More Meaningfully With Consumers. Regional Language Advertising Enables Brands To Communicate In A Language That Consumers Are Emotionally Attached To, Thereby Enhancing Comprehension, Trust, And Relatability.

Author: Mrs. R. Janani | Ms. R. Rupa Sri

Area of research: Education

Indian Students Are Increasingly Opting For Higher Education Outside India, Driven By Factors Such As Perceived Quality Of Education, Global Exposure, And Better Career Prospects. This Study Investigates The Perceptions Of Indian Students Towards Pursuing Higher Education Abroad, Including Their Motivations, Expectations, And Concerns. A Mixed-methods Approach Was Used To Collect Data From Indian Students Who Have Studied Or Are Currently Studying Abroad. The Findings Reveal That Indian Students Are Motivated By A Desire For Quality Education, Research Opportunities, And Global Exposure. However, They Face Challenges Such As Financial Constraints, Language Barriers, And Cultural Differences. The Study Provides Recommendations For Policymakers, Educators, And Institutions In India To Improve The Quality Of Higher Education And Develop Strategies To Retain Indian Students In The Country

Author: Dr W Saranya | M Divakar

Area of research: Natural Language Processing

The Interpretation Of Legal Documents Remains A Complex, Time-intensive Challenge For Both Legal Professionals And The General Public. This Paper Presents A Multilingual AI-Based Legal Document Analyzer That Lever- Ages Retrieval-Augmented Generation (RAG), Transformer- Based Natural Language Processing (NLP), And Multilingual Translation Models To Automate The Analysis Of Legal Con- Tracts And Agreements. The Proposed System Integrates A Clause Extraction Engine Built On Legal-BERT, A Semantic Question-answering Module Powered By FAISS-indexed Vector Retrieval And Flan-T5 Generation, A BART-based Document Summarizer, And A Multilingual Translation Pipeline Supporting English, Hindi, Tamil, And Telugu. Deployed Through An Interactive Streamlit Web Interface, The Platform Enables Users To Upload PDF Documents And Receive Real- Time Clause Highlights, Contextual Answers, Concise Sum- Maries, And Cross-lingual Translations. Experimental Evaluation On A Diverse Corpus Of Legal Documents Demonstrates Clause Extraction Precision Of 92%, Question-answering Ac- Curacy Of 88%, And Sub-1.5-second Response Latency, With 93% Of Survey Respondents Rating The Interface As Intuitive. The System’s Modular Architecture Supports Continuous Improvement Via Active Learning From User Feedback And Plug- And-play Model Upgrades.

Author: Dr. Arokiya Renjith | Avinash S | Raymond V | LohithRaaj A

Area of research: Electrical And Electronics Engineering

Road Accidents Often Result In Severe Consequences Due To Delays In Emergency Detection And Response. Conventional Systems Rely On Manual Reporting Or Single-sensor Mechanisms, Which May Lead To False Alerts And Slow Assistance. This Paper Proposes A Smart Accident Detection And Emergency Response System That Integrates Multi-sensor Monitoring With Real-time Communication And Intelligent Traffic Signal Control. The Vehicle-mounted Unit Utilizes A Vibration Sensor, Accelerometer, Gas Sensor, SpO₂ Sensor, And GPS Module To Detect Abnormal Conditions. A Timer-based User Confirmation Mechanism Is Incorporated To Minimize False Emergency Notifications. Upon Confirmation, Accident Data Along With Location Coordinates Are Transmitted Via Wi-Fi To A Mobile Application For Ambulance Coordination And Navigation. To Reduce Response Time Further, A Traffic Signal Preemption Mechanism Dynamically Assigns Green Signal Priority To The Ambulance Approach Lane, Overriding Normal Signal Cycles. The System Is Implemented Using An ESP8266- Based Architecture, Providing A Reliable And Cost-effective Solution For Real-time Accident Management And Smart Traffic Control.

Author: R. Vishwa Pandian | J. Alan Jones | S. Bharath | B. Tanushk | Dr.J.Devi Shree

Area of research: MARKETING

This Study Investigates The Influence Of Online Reviews On Students’ Shopping Behavior, Focusing On How Digital Feedback Shapes Purchasing Decisions. By Examining Patterns In Student Engagement With Product Ratings, Comments, And Recommendations, The Research Identifies Key Factors That Guide Their Trust And Choice In Online Marketplaces. Findings Reveal That Students Are Significantly Impacted By Both The Perceived Credibility And The Volume Of Online Reviews, Suggesting That Digital Word-of-mouth Plays A Pivotal Role In Shaping Consumer Behavior Within This Demographic. The Study Offers Insights For Marketers And E-commerce Platforms Aiming To Enhance Engagement And Tailor Strategies For Student Consumers.

Author: Ms. S. Subhitcha | Ms. C. Keerthana

Area of research: Artificial Intelligence And Data Science

The Rapid Growth Of Two-wheeler Usage Has Increased The Demand For Reliable Digital Platforms That Cater Specifically To The Needs Of Motorcycle Riders. However, Most Available Systems Focus Only On Isolated Services Such As E-commerce Or Service Booking, Lacking Integration And Rider-centric Features. The Proposed System, “Bikers Portal – One Stop Solution For Riders,” Aims To Develop A Centralized Digital Ecosystem That Integrates Service Booking, Spare Parts Purchasing, Maintenance Tracking, And Community Interaction Into A Single Web-based Platform. The System Enhances Rider Convenience By Providing Real-time Service Availability, Secure Authentication, Vehicle Profile Management, And Maintenance Reminders. Additionally, It Promotes Road Safety Awareness And Knowledge Sharing Through A Structured Community Forum. Built Using Modern Web Technologies And A Secure Database Framework, The Portal Ensures Scalability, Reliability, And Efficient Performance. The System Is Designed To Reduce Manual Effort, Improve Service Accessibility, And Strengthen Rider Engagement. Overall, The Bikers Portal Delivers A Comprehensive And Innovative Solution Tailored Exclusively For Motorcyclists.

Author: Roshan JP | Jagadeshan N | Jayaseelan R | Sanjay Kumar NA | Dr. Ananthi J (Guide)

Area of research: Artificial Intelligence

This Paper Presents The Development Of A Real-time Object Detection And Multi-object Tracking System Using The Ultralytics YOLO26 Model. The System Captures Live Video Through A Webcam And Processes Each Frame Using A Deep Learning-based Computer Vision Model To Detect Multiple Objects Simultaneously. The Detected Objects Are Highlighted With Bounding Boxes And Assigned Unique Tracking IDs To Maintain Their Identity Across Consecutive Frames. The System Is Implemented Using Python, OpenCV, And The PyTorch Framework. The Proposed Model Is Executed On CPU Hardware Without GPU Acceleration And Achieves An Average Processing Speed Of Approximately 18–25 Frames Per Second. Experimental Results Show That The System Can Detect Common Objects Such As Persons, Vehicles, And Everyday Items Effectively In Real Time. The Proposed Approach Provides A Cost-effective And Efficient Solution For Surveillance Systems, Smart Monitoring, And Intelligent Vision-based Applications.

Author: Reagan Joseph S | Santhosh B | Sabari Krishnan K | Ajina H

Area of research: B.com With Professional Accounting

This Study Investigates The Profound Socio-economic Shifts In Coimbatore City Driven By Recent Technological Integrations, Specifically Focusing On Industry 4.0, Artificial Intelligence And Smart City Initiatives. Traditionally Recognized As An Industrial And Textile Hub, Coimbatore Is Currently Undergoing A Rapid Digital Metamorphosis Into A Tier-2 IT Powerhouse. The Primary Objective Of This Research Is To Evaluate How These Technologies Have Altered Urban Living Standards, Industrial Productivity Among MSMEs, And Social Behavior Among The City’s Residents.

Author: Dr. Saranya W | Mr.AadithyaK

Area of research: Marketing

The Rapid Growth Of E-commerce Has Transformed The Way Businesses Operate And Consumers Shop. Meesho, As A Leading Social Commerce Platform In India, Relies On Innovative Marketing Strategies To Attract Customers, Enhance Brand Visibility, And Increase Sales In A Competitive Digital Marketplace. This Study Examines The Marketing Strategies Adopted By Meesho, Including Social Media Marketing, Reseller-based Promotion, Influencer Marketing, Pricing Strategies, Promotional Offers, Personalization, And Customer Relationship Management. Using Survey Data Collected From Online Shoppers, The Study Analyzes Consumer Perceptions Regarding The Effectiveness Of These Strategies. The Findings Indicate That Factors Such As Affordable Pricing, Reseller Recommendations, Attractive Discounts, And Strong Online Presence Significantly Influence Purchasing Decisions. The Study Concludes With Suggestions For Improving Marketing Effectiveness And Enhancing Customer Satisfaction.

Author: Dr W Saranya | Mr.E Mahath Nivan

Area of research: Information Technology

Automated Teller Machines (ATMs) Are Among The Most Common Banking Tools. They Mostly Rely On Physical Interaction Through Keypads And Touchscreens. This Can Raise Hygiene Concerns And Create Security Risks Like Shoulder Surfing And Keypad Tampering. This Paper Offers A Touchless ATM Access System That Combines Facial Recognition, Gesture-based Navigation, And A Virtual Keyboard For Secure Authentication And Transaction Processing. The System Applies The Haar Cascade Algorithm Along With Local Binary Pattern Histogram (LBPH) For User Identification, Followed By Password Confirmation Via A Gesture-based Virtual Keyboard. Hand Gesture Recognition, Powered By Computer Vision, Allows For Cursor Control And Menu Navigation Without Physical Contact. The Proposed Model Boosts Security With Multi-factor Authentication While Improving Accessibility And Hygiene By Removing Physical Touchpoints. Experiments Show Reliable User Authentication And Smooth Transaction Interaction With A Standard Webcam Setup. This Approach Provides A Practical And Scalable Framework For Future Contactless Banking.

Author: Harshitha D | Supraja M B | Thipirishetty Kavya | Dr. G. Ragu

Area of research: Artificial Intelligence And Data Science

The Rapid Growth Of Urbanization Has Significantly Increased Traffic Congestion, Posing Critical Challenges For Emergency Medical Services, Particularly Ambulances. Delays Caused By Conventional Fixed-time Traffic Signal Systems Can Lead To Increased Response Times And Adverse Outcomes For Patients. This Situation Highlights The Need For Intelligent, Automated, And Real-time Traffic Management Solutions Capable Of Prioritizing Emergency Vehicles. This Paper Presents An AI-Based Smart Traffic Control System For Emergency Vehicle Prioritization Using Deep Learning And Computer Vision Techniques. The Proposed System Employs A Custom-trained YOLOv8 Model To Accurately Detect Ambulances From Real-time Or Recorded Traffic Camera Feeds. The Detection Model Is Trained Using A Labeled Dataset Prepared With Roboflow, Ensuring High Precision For Ambulance-specific Classification. Upon Detecting An Ambulance With Sufficient Confidence, The Traffic Signal Dynamically Switches To A Green Phase, Allowing Uninterrupted Passage Through The Intersection. To Ensure System Stability And Prevent False Triggering, A Timeout-based Control Mechanism Is Incorporated, Enabling The Signal To Revert Safely To Normal Operation Once The Ambulance Exits The Camera’s Field Of View. A Graphical User Interface Is Developed To Visually Represent Traffic Signal States And Emergency Conditions, Providing Real-time Monitoring And Transparency. The System Is Implemented Using Python, OpenCV, And A Multithreaded Architecture To Maintain Real-time Performance Without Blocking The User Interface. Experimental Evaluation Demonstrates Reliable Ambulance Detection, Smooth Signal Transitions, And Effective Prioritization Under Varying Traffic Conditions. The Proposed Solution Enhances Emergency Response Efficiency, Reduces Manual Intervention, And Offers A Scalable Framework That Can Be Integrated Into Smart City Traffic Infrastructure. This Work Contributes Toward Improving Urban Emergency Mobility Through Intelligent Automation And AI-driven Traffic Management.

Author: Aishwarya

Area of research: Artificial Intelligence And Machine Learning For Wildlife Conservation

Illegal Wildlife Poaching Remains One Of The Most Critical Threats To Biodiversity Conservation, Leading To Ecological Imbalance, Species Extinction, And Economic Loss In Protected Reserves. Traditional Anti-poaching Strategies Rely Heavily On Manual Patrolling And Reactive Response Mechanisms, Which Are Inefficient In Large And Geographically Complex Forest Areas. There Is A Pressing Need For Intelligent, Data-driven Systems Capable Of Predicting High-risk Zones Before Poaching Incidents Occur. This Paper Presents WildGuard AI, A Wildlife Vulnerability Intelligence Engine Designed To Predict Poaching Risk Levels Across Protected Forest Zones Using Machine Learning Techniques. The Proposed System Utilizes Spatial, Environmental, And Historical Incident Data To Classify Forest Grids Into Low, Medium, And High-risk Categories. By Dividing Protected Reserves Into Structured 1 Km² Grids And Applying Supervised Learning Algorithms Such As Random Forest And Gradient Boosting, The System Generates Predictive Risk Heatmaps To Assist Forest Authorities In Strategic Patrol Deployment. The System Is Implemented Using Python, Flask Framework, And A Relational Database For Structured Storage Of Grid-level Intelligence Data. A Web-based Dashboard Visualizes Predicted Risk Zones, Patrol Allocation Data, And Vulnerability Metrics. Experimental Evaluation Demonstrates Reliable Classification Performance And Operational Feasibility. WildGuard AI Contributes Toward Proactive Conservation Strategies, Optimized Resource Allocation, And Technology-driven Wildlife Protection.

Author: Aadhikesavan D | John jabez A | Shanjay GS

Area of research: Artificial Intelligence And Healthcare Informatics, Focusing On Deep Learning-based Medical Image Analysis. It Explores The Integration Of Convolutional Neural Networks (CNNs) For Skin Lesion Classification With Large Language Models (LLMs) To Enhance Dia

Skin Diseases Are Among The Most Prevalent Medical Issues Globally, Making Timely And Accurate Diagnosis Essential For Successful Treatment.Expert Clinical Evaluation Is The Traditional Foundation Of Dermatological Diagnosis.However, This Approach Is Limited By Subjectivity, Accessibility, And Resource Availability. Automated Skin Disease Prediction From Medical Images Is Now Feasible Thanks To Recent Advances In Artificial Intelligence (AI) And Deep Learning. This Survey Comprehensively Examines Current Methods For Identifying Skin Diseases, Leveraging Machine Learning And Deep Learning. Specifically, It Reviews The Application Of Convolutional Neural Networks (CNNs),transfer Learning Models, And Integrated Hybrid AI Frameworks.The Analysis Covers The Strengths And Weaknesses Of Popular Classification Methods, Along With An Investigation Into Difficulties Like Imbalanced Datasets Lack Of Clarity, Poor Interpretability, And Problems With Generalization. This Survey's Core Contribution Is The Examination Of Integrated Frameworks That Combine CNN-based Image Classification With Large Language Models (LLMs). This Integration Facilitates The Delivery Of Explainable Diagnostic Insights, Stage Analysis, And Informed Treatment Recommendations. These Systems Bridge The Gap Between Automated Prediction And Clinical Decision Support By Integrating Visual Recognition With Semantic Understanding

Author: Bharathwaj R | Jayasimma D | Malan E K | Mr. B. Sathishkumar

Area of research: MARKETING

This Study Focuses On Understanding The Strategic Objectives Aimed At Building A Strong And Sustainable Business Model. The Primary Objective Is To Deliver A Seamless Ecosystem Experience That Integrates Products, Services, And Customer Interactions Into A Unified And Convenient Platform. The Study Also Emphasizes Maintaining High Product Quality And Continuous Innovation To Meet Evolving Consumer Expectations And Remain Competitive In The Market. Furthermore, It Highlights The Importance Of Creating A Strong Brand Identity And Building Customer Trust, Which Are Essential For Long-term Loyalty And Business Growth. By Analyzing These Objectives, The Study Aims To Explore How An Organization Can Enhance Customer Satisfaction, Strengthen Market Positioning, And Achieve Sustainable Development Through Quality, Innovation, And Trust-based Relationships.

Author: Dr.Vadivel M | Thiruselvam M

Area of research: MARKETING

This Study Examines Nykaa’s Marketing Mix And Its Effect On Consumer Choices In The Competitive Beauty And Personal Care Industry. Nykaa Has Emerged As One Of India’s Leading Online And Offline Beauty Retailers By Effectively Implementing The 4Ps Of Marketing—Product, Price, Place, And Promotion. The Research Analyzes How Nykaa’s Wide Product Range, Competitive Pricing Strategies, Strong Online Presence, And Influencer-driven Promotional Campaigns Influence Consumer Purchasing Decisions. The Study Also Explores Factors Such As Brand Trust, Product Authenticity, Convenience, Discounts, And Personalized Recommendations. Data Collected Through Surveys And Secondary Sources Interpretation Primary Data Indicate That Digital Marketing, Social Media Engagement, And Customer-centric Strategies Significantly Impact Consumer Preferences And Brand Loyalty. The Findings Suggest That Nykaa’s Well-integrated Marketing Mix Plays A Crucial Role In Shaping Consumer Perceptions And Driving Repeat Purchases. This Study Provides Insights Into How Strategic Marketing Decisions Influence Buying Behavior In The Growing Online Skincare And Cosmetics Market.

Author: Dr.Vadivel M | S.Sowmiya

Area of research: MARKETING

This Research Investigates How Cultural Adaptation Influences Consumer Trust Toward KFC In India. While Initial Interactions With The Brand Are Largely Driven By Curiosity And Global Appeal, Sustained Acceptance Depends On Cultural Comfort And Perceived Reliability. The Study Analyses How Localized Menus, Culturally Sensitive Practices, And Consistent Service Quality Contribute To Trust-building Among Consumers. Primary Data Were Collected From 120 Respondents Through A Structured Questionnaire And Analysed Using Percentage Analysis, Ranking Analysis, And Likert Scale–based Weighted Averages. The Results Demonstrate That Cultural Familiarity And Hygiene Perception Significantly Enhance Consumer Confidence And Repeat Patronage. The Study Concludes That KFC’s Success Lies In Its Ability To Integrate Global Branding With Local Cultural Expectations, Transforming Curiosity-driven Trials Into Long-term Consumer Comfort.

Author: Harish V | Dr D Sharon

Area of research: MARKETING

Food Habits Among College Students Are Undergoing A Rapid Transformation, Shaped By Academic Pressure, Urban Lifestyles, Technological Influence, And Increasing Independence In Food Choices. This Transitional Phase Of Life Marks A Critical Period Where Young Adults Move Away From Home-cooked Meals Toward Convenience-based And Commercially Prepared Foods. The Present Study Explores The Evolving Dietary Patterns Of College Students, Focusing On Meal Regularity, Food Preferences, Nutritional Awareness, And The Impact Of Social And Environmental Factors. It Highlights How Factors Such As Time Constraints, Peer Influence, Digital Food Delivery Platforms, And Stress Contribute To Irregular Eating Habits And Nutritional Imbalance. While Students Increasingly Prioritize Taste, Affordability, And Convenience, Health Considerations Often Take A Secondary Role. The Study Emphasizes The Long-term Implications Of These Habits On Physical Health, Mental Well-being, And Academic Performance. By Examining The Changing Relationship Between Young Minds And Their Food Choices, This Research Aims To Draw Attention To The Need For Nutritional Awareness And Institutional Support To Encourage Healthier Eating Practices Among College Students.

Author: Dr W Saranya | Ms R Arbutha

Area of research: Intelligent Transportation Systems

Urbanization Has Led To A Rapid Increase In Vehicle Density, Resulting In Frequent Road Congestion And Accidents. According To Global Traffic Studies, A Significant Percentage Of Urban Accidents Occur Due To Delayed Reaction Time, Blind Spots, And Lack Of Situational Awareness. Advanced Driver Assistance Systems (ADAS) Have Been Introduced To Reduce Accident Rates. These Systems Use Hardware Sensors Such As LiDAR And Radar To Detect Obstacles And Nearby Vehicles. Although Effective, These Systems Are Costly And Mostly Limited To High-end Vehicles. In Contrast, Smartphones Have Become Ubiquitous And Contain Advanced Sensing And Communication Capabilities. These Devices Can Continuously Monitor Location, Velocity, Orientation, And Motion Patterns. Leveraging These Built-in Sensors For Collaborative Vehicle Safety Presents An Affordable Alternative To Traditional Hardware-dependent Systems. The Primary Objective Of This Research Is To Design And Implement A Low-cost Smartphone-based Collaborative System That Enhances Urban Vehicle Safety Without Requiring Additional Hardware Installations.

Author: Srikumaran S | Sam Stephen S | Sanjai G | Ranjith Kumar S | PRADEEP K (Mentor)

Area of research: Artificial Intelligence And Machine Learning In Healthcare

The Rapid Advancement Of Artificial Intelligence (AI) And Machine Learning (ML) Technologies Has Significantly Transformed Various Domains, Particularly The Healthcare Sector. Early Disease Detection And Timely Medical Consultation Play A Crucial Role In Reducing Mortality Rates, Treatment Costs, And Healthcare Burden. However, Many Individuals Delay Hospital Visits Due To Lack Of Awareness, Limited Accessibility, Long Waiting Times, Or Uncertainty Regarding The Severity Of Symptoms. This Gap Highlights The Need For Intelligent, Accessible, And Integrated Healthcare Assistance Systems. This Paper Presents A Smart Healthcare Disease Prediction And Appointment System Using Machine Learning, A Web-based Application Designed To Provide Preliminary Medical Diagnosis Support And Seamless Doctor Appointment Scheduling Within A Unified Platform. The Proposed System Leverages Supervised Machine Learning Algorithms Trained On Structured Symptom-disease Datasets To Predict Possible Diseases Based On User-input Symptoms And Demographic Attributes Such As Age And Gender. The Prediction Engine Computes Probability Scores For Multiple Disease Classes And Identifies The Most Probable Condition Along With A Confidence Percentage. Additionally, The System Displays The Top Three Predicted Diseases To Enhance User Awareness And Informed Decision-making. To Further Enhance Reliability, The System Incorporates A Risk Classification Mechanism That Categorizes Predicted Diseases Into Low, Medium, And High Risk Levels Based On Model Confidence And Disease Severity. Unlike Traditional Symptom Checker Applications That Provide Only Textual Suggestions, The Proposed Platform Integrates A Complete Appointment Management Module. Users Can View Available Doctors Categorized By Specialization, Book Appointments Directly After Prediction, And Maintain Structured Appointment History Records. The System Is Implemented Using Python, Flask Framework, And A Relational Database For Secure Data Storage And Session Management. The Modular Architecture Ensures Scalability And Extensibility For Future Integration With Hospital Management Systems, Chatbot Assistance, And Mobile Applications. Experimental Evaluation Demonstrates Reliable Prediction Performance And Smooth System Functionality. The Proposed Solution Contributes Toward Enhancing Digital Healthcare Accessibility, Reducing Unnecessary Hospital Visits, And Supporting Early-stage Medical Decision-making Through Intelligent Automation.

Author: JanarthananV | JananiPriyaV | ShamlinThisha J | Abirami R

Area of research: Artificial Intelligence And Smart Agriculture

Crop Damage Caused By Wild Animals, Birds, And Unauthorized Human Intrusion Remains A Major Challenge In Modern Agriculture. Traditional Monitoring Approaches Such As Manual Supervision, Fencing, And Static Alarm Systems Are Often Inefficient, Labor-intensive, And Unable To Provide Continuous Protection. This Paper Presents Agri-Guard, An AI-powered Smart Agricultural Surveillance System That Leverages Real-time Computer Vision And Deep Learning For Automated Detection And Deterrence Of Threats In Farmland Environments. The Proposed System Utilizes The YOLOv8 Object Detection Model For Identifying Humans, Animals, And Birds From Live Camera Feeds. Video Streams Are Processed Using OpenCV, And Detections Are Handled Through A FastAPI-based Backend Architecture. Upon Detecting A Threat, The System Triggers Intelligent Alert Mechanisms, Records Evidence, And Logs Detection Data Into A Structured Database For Monitoring And Analysis. Experimental Evaluation Demonstrates That The System Achieves High Detection Accuracy With Low Latency, Making It Suitable For Real-time Agricultural Deployment. The Modular Architecture Ensures Scalability, Hardware Extensibility, And Integration With IoT-based Deterrent Mechanisms Such As Lights, Alarms, And Laser Systems.

Author: DHARUN A | AKASH SARAVANA SINGH R | REDONE RAJ S | AKASH G | Dr P Shenbagavalli (GUIDE)

Area of research: Electronics And Communication Engineering

Lung Cancer Is One Of The Most Life-threatening Diseases Worldwide, And Early Diagnosis Significantly Increases The Chances Of Survival. Accurate Detection And Classification Of Lung Tumors Remain Challenging Due To Variations In Tumor Size, Shape, And Texture In MRI Images. This Study Proposes An Automated Lung Tumor Segmentation And Classification Framework Using Convolutional Neural Networks (CNN) And The Visual Geometry Group (VGG) Network Architecture. The Proposed System Utilizes MRI Images To Analyze Textural And Spatial Features Of Lung Tissues For Distinguishing Between Normal And Malignant Cases. A Multi-scale Feature Extraction Approach Is Incorporated To Improve Detection Performance And Enhance Classification Accuracy. The VGG Network Serves As The Base Model For Deep Feature Learning, While CNN Layers Refine Segmentation And Classification Tasks. The Developed Database Includes Multiple MRI Views To Ensure Robust Training And Validation. Experimental Results Demonstrate That The Proposed Model Achieves High Precision And Overall Classification Accuracy Of Up To 98%, As Evaluated Using Confusion Matrix Metrics. The System Reduces Manual Interpretation Errors And Provides An Efficient Computer-aided Diagnostic Tool For Early Lung Tumor Prediction.

Author: Mrs.M.Geethalakshmi, Nithish S | Nithish S | Gokul K

Area of research: Marketing

Television Advertising Plays A Significant Role In Shaping Consumer Buying Behavior By Influencing Awareness, Attitudes, And Purchase Decisions. This Study Examines The Effect Of Television Advertisements On Consumers By Analyzing How Visual Appeal, Message Content, Repetition, Celebrity Endorsements, And Emotional Elements Impact Buying Intentions. Television Advertisements Act As A Powerful Communication Tool That Not Only Informs Consumers About Products And Services But Also Persuades Them By Creating Brand Recall And Positive Perceptions. The Study Highlights That Frequent Exposure To Television Advertisements Increases Product Familiarity And Trust, Which In Turn Affects Consumers’ Preferences And Choice Of Brands. Moreover, Advertisements Targeting Emotions And Lifestyles Are Found To Be More Effective In Motivating Impulse Buying And Brand Loyalty. The Findings Suggest That Television Advertising Continues To Be An Influential Factor In Consumer Decision-making, Despite The Growth Of Digital Media, And Remains A Crucial Strategy For Marketers To Attract, Influence, And Retain Consumers In Competitive Markets.

Author: Dr W Saranya | Mr. B L Santosh

Area of research: Information Technology

Email Attachments Remain A Primary Vector For Phishing And Malware Attacks, With Traditional Signature-based Defences Struggling Against Zero-day Threats. This Paper Introduces Opaline, A Proactive Defence System That Leverages Deep Learning Models Like RoBERTa To Analyse Textual Semantics And Structural Features In Attachments Such As PDFs And Documents, Achieving Early Detection Before User Interaction. By Integrating Sandbox Isolation And Automated Sanitization—converting Suspicious Files Into Safe Static Previews—Opaline Minimizes Risks While Preserving Workflow Usability, Offering A Practical Advancement Over Resource-heavy Existing Solutions.

Author: Mrs. M.Radhika | G.Ashika | Shreya Agrawal | T.Sruthik

Area of research: Machine Learning

Customer Retention Has Become A Major Concern For Subscription-based B2B Software As A Service (SaaS) Companies Because Long-term Contracts Directly Influence Revenue Stability. Even A Small Increase In Churn Rate Can Result In Noticeable Financial Loss For Service Providers. In This Study, Machine Learning Techniques Are Applied To Analyze And Predict Customer Churn Using Structured Enterprise Data. The Proposed Framework Includes Data Preprocessing, Handling Class Imbalance Using The Synthetic Minority Oversampling Technique (SMOTE), And Evaluating Several Ensemble Learning Models Including Random Forest, AdaBoost, CatBoost, And LightGBM. The Models Are Assessed Using Commonly Used Classification Metrics Such As Accuracy, Precision, Recall, F1-score, And ROC-AUC. Special Attention Is Given To Recall Since Identifying Potential Churn Customers Is Particularly Important For Business Decision-making. Among The Evaluated Approaches, LightGBM Demonstrated Stable And Comparatively Better Performance Across Multiple Metrics. To Improve Transparency, SHAP Analysis Is Used To Interpret Feature Contributions And Identify Factors Influencing Churn Behavior. The Trained Model Is Further Integrated Into A Streamlit-based Dashboard That Allows Real-time Predictions And Batch Processing Of Customer Data, Helping Organizations Monitor Churn Risk And Plan Proactive Retention Strategies.

Author: Minu Ashika A | Azhagu Meena P | Dr. Saranya | Dr. J. Arokia Renjit