Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Sivaram Ponnusamy has a total of 74 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Dark Web Crime Intelligence as a Predictor of Crime Through AI, Rhythm Analytics, and Sonic Forecasting

    by Hariharan Kalyanaraman, Deemed To Be, Sivaram Ponnusamy, A. L. Sriram

    Dark-web criminality now assumes the form of a constantly dynamic, high-velocity ecosystem that places significant stress on traditional, stagnant and responsive crime-intelligence paradigms. This chapter describes my project, a rhythm-based predictive intelligence model that turns Dark-web criminal behaviour into a signal of time, as opposed to a disaggregated dataset. The suggested solution enables the identification of pre-crime patterns in the early-stage by combining the concepts of artificial intelligence, beat modelling, tempo -acceleration analysis and frequency-domain anomaly detection. These resilient, active risk pathways and prospective decision-making cycles are supported by the framework based on the signal-processing theory, music-inspired temporal analogies and cognitive virtues of sonic abstraction

  • Artificial Intelligence for Biomarker Discovery and Clinical Trial Optimization

    by S. Gejalakshmi, T. Sriramya, S. Salma Shaik, N. Harikrishnan, C. Aswin, K. Dharanitharan, Sivaram Ponnusamy

    Biomarkers exist in various forms including molecular and genetic elements as well as imaging techniques and physiological and behavioral and digital markers which function as measurable indicators to track biological processes and disease states and therapeutic responses. These elements establish the basis for precision medicine through their capacity to detect diseases at their earliest stages and to classify risks and disease types and to track treatment results in real-time. Hence plays crucial role in AI field

  • Deep Learning-Driven Pothole Dimension Analysis for Prioritizing Road Maintenance

    by Sachin Bajirao Bhosale, Sivaram Ponnusamy

    The impacts of potholes on road safety and maintenance costs are immense, hence requiring efficient identification and prioritization methods. This paper provides an extensive framework for integrating CNNs with GIS tools to automate pothole dimension analysis. It takes road surface images, feeds them into the CNN model to accurately detect and measure potholes, and visualizes spatial distribution using GIS tools, thus enabling data-driven prioritization for repair. This approach enhances decision-making through precise metrics on pothole size, depth, and location, optimizing resource allocation. The framework demonstrates high accuracy and scalability, offering a transformative solution for proactive road infrastructure management.

  • AI-Enhanced Wearable Technologies

    by Shubhangi Tulshidas Dhamdhere, Sivaram Ponnusamy

    Smart glasses that include Artificial Intelligence, Augmented Reality, and live object detection have been engineered to ensure safety, security, and independence for women. Based on the YOLO algorithm, the glass allows users to accurately perceive objects, and the technology provides audio feedback to safely guide the user around any potential dangers. Personal safety issues are also considered. It is equipped with a voice-activated SOS alarm, which immediately sends the GPS location of the user to the emergency contact using GSM technology. In addition, the health sensors in this equipment continuously monitor the vital signs and send alerts automatically. It features a built-in AI-powered virtual assistant with GPT integration, which enables hands-free messaging, weather updates, and notification management. These smart sunglasses bring forth an innovative wearable solution that empowers women, enhancing their independence and overall security in daily life by combining cutting-edge technology for safety, health monitoring, and digital assistance.


  • AI-Driven E-Court Systems in India

    by Shyamrao A. Gade, Sivaram Ponnusamy

    The incorporation of AI into India's E-Court systems presents a significant opportunity to improve the efficiency, accessibility, and reliability of judicial operations. This research study concentrates on designing and developing an AI-driven E-Court system by collecting and analyzing legal information, including Indian Penal Codes (IPC), constitutional provisions, court opinions, legal briefs, statutes, and case law. The proposed system will utilize sophisticated data retrieval and natural language processing (NLP) methods to enhance legal research, automate document generation, and forecast case results. The gathered datasets will undergo pre-processing for consistency, be labeled for legal entities, and be enhanced with metadata to facilitate powerful machine learning (ML) models. It aligns with the constitutional mandate of delivering timely justice under Article 21 (Right to Life and Personal Liberty) and enhancing the accessibility of justice under Article 39A (Equal Justice and Free Legal Aid). The findings highlight how AI can transform judicial processes in India.

  • Federated LSTM Framework for Distributed Traffic Forecasting and Mobility Optimization in Blockchain-Integrated VANETs

    by K. Swaminathan, Sivaram Ponnusamy, R. Karthikeyan

    This study presents a Federated LSTM framework designed to improve traffic forecasting and mobility optimization in blockchain-integrated Vehicular Ad Hoc Networks (VANETs). By leveraging a federated learning approach, the framework enables collaborative model training across decentralized nodes without requiring direct data sharing, enhancing privacy and data security among network participants. The LSTM model effectively captures temporal traffic patterns, providing accurate and real-time forecasting to support dynamic mobility management. The integration of blockchain ensures data integrity and facilitates secure, transparent communication between vehicles and infrastructure within the VANET. This decentralized approach addresses the challenges of data privacy, system scalability, and efficient resource utilization in traffic prediction and mobility control. The proposed framework holds promise for advancing smart transportation systems by reducing congestion, improving traffic flow, and enhancing the overall efficiency of urban mobility

  • Secure Data Exchange in VANETs Using Blockchain and Support Vector Machines for Smart Urban Mobility

    by R. Karthikeyan, K. Swaminathan, Sivaram Ponnusamy

    This research proposes a machine learning-based framework integrating blockchain technology and Support Vector Machines (SVM) to secure data exchange in Vehicular Ad Hoc Networks (VANETs) for smart urban mobility. VANETs, essential to modern smart city infrastructure, enable vehicle-to-vehicle and vehicle-to-infrastructure communication, facilitating real-time information sharing. However, these networks face significant security and privacy challenges due to their open and dynamic nature. By employing blockchain, the framework ensures data integrity and prevents unauthorized access, creating a secure environment for data transactions within VANETs. The SVM algorithm enhances data processing efficiency by accurately classifying and filtering data exchanged across the network, mitigating potential threats and minimizing vulnerabilities.


  • Deep Federated Learning for Real-Time Decision Making in Autonomous Vehicles

    by Vaibhav Dhage, Sivaram Ponnusamy

    Autonomous vehicles (AVs) rely on real-time decision-making to navigate dynamic environments. As the complexity of driving scenarios grows, traditional machine learning techniques face limitations in ensuring reliable, scalable, and secure decision-making. This chapter explores the integration of deep federated learning (DFL) to address these challenges by enabling distributed, privacy-preserving, and collaborative learning among AVs. We propose a novel DFL model tailored to improve real-time decision-making processes in AVs, leveraging vehicle data from various sources while maintaining data privacy. The chapter covers the theoretical foundations, design, and implementation of this model, with experimental results showcasing significant improvements in decision accuracy, latency reduction, and resource efficiency. The proposed system's advantages, social welfare implications, and potential for future enhancements are also discussed.