M. Suganiya

@srmist.edu.in

Assistant Professor, Department of Computing Technologies
SRM Institute of Science and Technology

M. Suganiya


                  Download Compact PDF  
https://researchid.co/suganiyam

RESEARCH INTERESTS

Artificial Intelligence, Machine Learning, Image Processing
24

Scopus Publications

Scopus Publications

  • Machine learning in law: unveiling the AHKPSV algorithm for improved legal case outcome forecasts
    Vaissnave Venkadapathi, Suganiya Murugan, Selva Birunda Shanmugavel, Dharani Veeriaya, Revathi Balaganapathy
    International Journal of Machine Learning and Cybernetics, 2026
  • AVATAR-MUSE: An Automated, Avatar-Based Chatbot System for Museum Ticket Booking
    Pranay Jha, S. Pradeep Kumar, S. Manoj, A. Sinduja, Suganiya Murugan
    Proceedings of 2nd International Conference on Multi Agent Systems for Collaborative Intelligence Icmsci 2026, 2026
    The booking of museum tickets is a critical yet often cumbersome process for cultural enthusiasts, tourists, and casual visitors alike. Conventional web portals for museum reservations are frequently text-heavy, cluttered, and unintuitive, leading to poor user experience and decreased accessibility. This paper presents AVATAR-MUSE, a next-generation, automated, webbased system that revolutionizes the ticket booking process by combining an intelligent conversational AI with a visually engaging, avatar-driven interface. Leveraging advanced Natural Language Processing (NLP) and deep learning models, AVATARMUSE interprets diverse user inputs, understands conversational context, and responds with relevant booking information in real-time. Unlike static booking websites, AVATAR-MUSE incorporates a human-like avatar, powered by WebGL-based rendering and motion animation, to guide users through the process in a friendly and intuitive manner. The system is tightly integrated with museum APIs to fetch live data about exhibitions, available time slots, ticket prices, and booking statuses. Additionally, it supports secure, real-time payment processing and booking confirmation, ensuring a seamless end-toend user journey. By merging voice/text input processing, secure backend architecture, and visually immersive interaction, AVATAR-MUSE redefines how users engage with cultural institutions online. This paper details the architecture, implementation, and evaluation of the system and highlights its potential to be extended across other domains such as theaters, concerts, and heritage tourism.
  • VISIPEDIA-Automated Web-based System for Chemical Compound Analysis
    Akarsh Singh, Sampurna Saha, Suganiya Murugan
    Proceedings 3rd International Conference on Self Sustainable Artificial Intelligence Systems Icssas 2025, 2025
    The analysis of chemical compounds plays a crucial role in fields ranging from education to drug discovery, yet accessing detailed molecular information typically requires specialized software and domain expertise. This paper introduces VISI-PEDIA, an automated, web-based system designed to simplify chemical compound analysis by focusing on essential molecular attributes. VISIPEDIA combines deep learning, image processing, and web technologies to provide a user-friendly, accessible tool for molecular data retrieval. By integrating the Flask web framework with RDKit, a cheminformatics library, VISIPEDIA enables users to input compound names and receive immediate, structured data output, including molecular formula, chemical weight, number of rings, and IUPAC (International Union of Pure and Applied Chemistry) name. Testing on 50 benchmark compounds showed 75 % accuracy in SMILES conversion and property extraction. Additionally, VISIPEDIA incorporates image-based molecular recognition powered by deep learning models, including Convolutional Neural Networks (CNNs) and ResNet18, to support structure analysis from molecular images, achieving 92.7% classification accuracy. This hybrid approach allows the platform to process and interpret both text-and image-based chemical data, making it versatile for educational and research applications. User feedback from 20 participants highlighted system usability and response time, affirming its utility. The system’s design emphasizes accessibility, empowering students, educators, and researchers to engage in molecular analysis without requiring extensive background knowledge. Potential applications of VISIPEDIA span academic research, chemistry education, and preliminary drug discovery research, establishing it as a scalable and impactful solution in the field.
  • Greenhouse Environment Monitoring Using Raspberry Pi
    Ritam Biswas, Aman Pandey, Suganiya Murugan
    Proceedings of International Conference on Visual Analytics and Data Visualization Icvadv 2025, 2025
    With the increasing emphasis on sustainable agriculture, there is a need to stress on technological advancement which can optimize the growth of plants under controlled conditions. We are proposing here an automated greenhouse monitoring system having a low cost which makes sure that parameters like temperature, humidity, soil moisture and the light intensity are constantly supervised by the Automatic Control. The specific system described in this paper includes several sensors and is a completely automated, cost-effective Raspberry Pi Microcontroller. Use the Internet (IoT) of things to provide remote monitoring and management so that users can access live data so that they can adjust anywhere. All agricultural processes are dealt with responsible management due to automation so there is a practical move towards targeted assistance. The greenhouse design which was drawn looking at the working life was aimed at suiting different climatic conditions and sustaining the various culture cycles. This is meant to encourage sustainable farming by providing a cheap and simple solution that improves greenhouse management. The system achieves high accuracy in monitoring key parameters, with temperature and humidity measurements accurate up to 98%, while keeping constant real-time visuals on the parameters. This stable environment positively impacts plant productivity while promoting responsible resource conservation. The results demonstrate the practical potential of IoT in agriculture, highlighting its application in enhancing productivity and sustainability.
  • CareChain - Transforming Health Data Security with Blockchain and Distributed Storage
    Yohesh R, Tejashree N, Suganiya Murugan
    2025 6th International Conference on Data Intelligence and Cognitive Informatics Icdici 2025, 2025
    CareChain implements EMR storage through a decentralized system that uses IPFS with Ethereum blockchain infrastructure. CareChain provides healthcare institutions with an alternative solution by addressing data risks inherent in traditional centralized management systems without patient control. Off-chain storage with IPFS works efficiently and at scale, just as smart contracts make it possible to control medical information access. Patients can manage access to trusted healthcare professionals by giving them control over their medical records through the system. Quick and accurate medical information exchange through this system helps healthcare practitioners improve diagnosis and therapy. By being decentralized, the platform delivers better security along with stronger reliability by breaking its dependence on one central authority. The CareChain platform provides safe medical information exchange between different healthcare systems while simultaneously ensuring reliable data privacy and accuracy. The system architecture designed for patients enables secure handling and sharing of medical data through a dependable structure that grows with demand, fitting digital healthcare trends.
  • Optimizing Machine Learning Models with CUDA: A Comprehensive Performance Analysis
    Niteesh L, Ampareeshan M B, Suganiya Murugan
    2025 3rd International Conference on Communication Security and Artificial Intelligence Iccsai 2025, 2025
    Deep neural networks (DNNs), a subset of machine learning models, often face training and inference computational bottlenecks that demand extensive computational resources. These challenges are uniquely addressed through acceleration on Graphics Processing Units (GPUs). However, fully exploiting GPU architecture requires carefully pipelined tasks and low-level optimizations. This paper investigates how Compute Unified Device Architecture (CUDA) can significantly enhance the performance of machine learning models by optimizing critical computational kernels, memory management, and parallelization strategies. We demonstrate substantial reductions in execution time and improved resource utilization through CUDA techniques such as memory coalescing, shared memory usage, and kernel fusion. Our benchmarks reveal up to a 3.65x speedup in matrix operations and a 2.32x increase in CNN training throughput, establishing CUDA optimization as a practical solution for modern, high-efficiency machine learning workloads. These results underscore the importance of low-level GPU optimization in enabling scalable and energy-efficient AI systems. Finally, we offer guidelines for researchers and practitioners to effectively leverage CUDA to accelerate machine learning tasks and bridge the gap between high-level frameworks and hardware capabilities.
  • Comparison of Machine Learning Models for Injury Prediction in Athletes
    Suganiya Murugan, S. Pradeep Kumar, Kalaiarasi G, B. Saritha, B. Rubini
    Proceedings of International Conference on Visual Analytics and Data Visualization Icvadv 2025, 2025
    Running is a favoured exercise for millions, but it carries a risk of musculoskeletal injuries, particularly to the lower extremities. Identifying and predicting potential injuries is crucial for prevention. This study explores the use of machine learning techniques to predict injuries among runners, leveraging data on demographics, running habits, health conditions, and previous injuries. Using models like Random Forest, K-Nearest Neighbours, Decision Trees, and Logistic Regression, the study examines the relevance of various attributes and assesses the predictive power of these models. Remarkably, our model achieved an impressive accuracy of over 99%, setting a new benchmark for predictive precision in injury prevention. Visualizations provide insight into attribute interactions and their influence on injury risk, facilitating the development of targeted prevention strategies.
  • Computer Aided Diagnosis Multi-model System using Late Fuison and Ensemble Learning
    M.Kani Priya, Gunabhiram Aruru, Nikhil Vanam, M. Suganiya, M. Salomi
    2025 International Conference on Pervasive Computational Technologies Icpct 2025, 2025
    Lung Cancer and brain tumors are the well-known causes of cancer deaths universal. Therefore, appropriate and accurate diagnosis is an important issue that affects better and more reliable treatment and the patient's chance of survival. Present diagnostic strategies are mostly based on a single piece of information. This often leads to missing necessary information, which can lead to Misidentification of condition or delay in timely treatment. This study presents a novel Computer-Aided-Diagnosis system designed to enhance the detection of lung cancer and brain tumors by integrating multimodal data, including medical images (CT and MRI scans) and clinical information. The proposed approach employs late fusion and ensemble learning techniques, utilizing CNN for processing CT images, VGG19 for analyzing MRI scans, and Random Forest classifiers for evaluating clinical data. By separately training these models, the system capitalizes on the distinct characteristics of each data modality, overcoming the limitations of traditional CAD systems that typically rely on a single data source. This comprehensive analysis not only improves diagnostic accuracy but also provides reliable predictions, ultimately facilitating better clinical decision-making for lung cancer and brain tumor detection.
  • Credit Card Fraud Detection Using Machine Learning
    Pratap Singh Raj, Suganiya Murugan, M. Kanipriya
    Aip Conference Proceedings, 2024
  • Automated detection of credit card fraud through machine learning
    Mayank Shekhar, Krishna Khawas, Suganiya Murugan
    Aip Conference Proceedings, 2024
  • Interactive Sign Language Learning System Using Computer Vision and Deep Learning
    Suganiya Murugan, Mir Kasif Ali, Dhanvanth Singari, S. Pradeep Kumar
    Proceedings 3rd International Conference on Advances in Computing Communication and Applied Informatics Accai 2024, 2024
  • Predictive Modeling for Asset Bubble Detection in Financial Markets
    Suganiya Murugan, Pradeep Kumar Sivakumar, Arpit Goyal, Pranay Goenka
    International Conference on Smart Systems for Electrical Electronics Communication and Computer Engineering Icsseec 2024 Proceedings, 2024
  • Next - Gen Driver Monitoring System
    Suganiya Murugan, Harish Krishnadas S, Abiishek G, Pradeep Kumar S, Rubini B
    Proceedings 2024 IEEE 6th Global Power Energy and Communication Conference Gpecom 2024, 2024
  • Performance Analysis of Insulation Resistance Test And Cable Laying Installation In Library Block
    B. Rubini, R. Karthick, Kalaiarasi G, S. Saranya, S. Pradeep Kumar, M. Suganiya
    Proceedings 2024 IEEE 6th Global Power Energy and Communication Conference Gpecom 2024, 2024
  • Filter Duplicate Images using CLIP
    Diptayan Jash, Tuhina Tripathi, Kanupriya Johari, Avya Rathod, Gayathri M, Suganiya M, Ankit Mishia, Athira S Menon
    7th International Congress on Ambient Intelligence Software Engineering and E Health and Mobile Health Amitic 2024, 2024
  • An Electro-Oculogram (EOG) Sensor’s Ability to Detect Driver Hypovigilance Using Machine Learning
    Suganiya Murugan, Pradeep Kumar Sivakumar, C. Kavitha, Anandhi Harichandran, Wen-Cheng Lai
    Sensors, 2023
  • A Comparison of COVID-19 Detection using Deep Learning Methods
    S. Pradeep Kumar, Suganiya Murugan, B. Rubini
    Proceedings of the 2023 2nd International Conference on Electronics and Renewable Systems Icears 2023, 2023
  • A Machine Learning Approach to Predict Skin Diseases and Treatment Recommendation System
    Suganiya Murugan, S.R. Srividhya, S. Pradeep Kumar, B. Rubini
    Proceedings 5th International Conference on Smart Systems and Inventive Technology Icssit 2023, 2023
  • On-site Testing and Preventive Maintenance for a Fire Automation System Employing PSO, FA and ANN Algorithms
    B. Rubini, G Kalaiarasi, S. Vasuki, S. Pradeep Kumar, Suganiya Murugan, C. Arul Stephen
    1st International Conference on Emerging Research in Computational Science Icercs 2023 Proceedings, 2023
  • Implementation of a Smart Garbage Monitoring System using GSM
    B. Rubini, S. Pradeep Kumar, M. Suganiya
    International Conference on Automation Computing and Renewable Systems Icacrs 2022 Proceedings, 2022
  • Detection of SQL Injection Attack Using Adaptive Deep Forest
    M.S. Roobini, S.R. Srividhya, Sugnaya, Kannekanti Vennela, Guntumadugu Nikhila
    2022 International Conference on Communication Computing and Internet of Things Ic3iot 2022 Proceedings, 2022
  • Detection and analysis: driver state with electrocardiogram (ECG)
    Suganiya Murugan, Jerritta Selvaraj, Arun Sahayadhas
    Physical and Engineering Sciences in Medicine, 2020
  • Driver Hypovigilance Detection for Safe Driving using Infrared Camera
    Suganiya Murugan, Jerritta Selvaraj, Arun Sahayadhas
    Proceedings of the 5th International Conference on Inventive Computation Technologies Icict 2020, 2020
  • Analysis of different measures to detect driver states: A review
    Suganiya Murugan, Jerritta Selvaraj, Arun Sahayadhas
    2019 IEEE International Conference on System Computation Automation and Networking Icscan 2019, 2019