Ph. D. in Computer Science and Engineering (Area of Spe-ML) Dec, 2015 – Jul, 2020
CHRIST (Deemed to be University), Bengaluru, Karnataka, India.
Masters in Software Engineering Aug, 2008 – Dec, 2010
Jawaharlal Nehru Technological University, Kakinada, India
Bachelors in Information Technology Aug, 2004 – April, 2007
Andhra University, Visakhapatnam, India
RESEARCH, TEACHING, or OTHER INTERESTS
Computer Engineering, Artificial Intelligence, Computer Science Applications, Information Systems
Categorizing Mental Stress: A Consistency-Focused Benchmarking of ML and DL Models for Multi-Label, Multi-Class Classification via Taxonomy-Driven NLP Techniques Juswin Sajan John, Boppuru Rudra Prathap, Gyanesh Gupta, Jaivanth Melanaturu Natural Language Processing Journal, 2025 Mental stress, a critical concern worldwide, necessitates precise and nuanced characterization. This study introduces a novel approach to effectively characterize mental stress through a multi-label, multi-class classification framework through natural language processing techniques. Building on existing literature, discussions with psychologists and other mental health practitioners, we developed a taxonomy of 27 distinctive markers spread across 4 label categories; aiming to create a preliminary screening tool leveraging textual data. The core objective is to identify the most suitable model for this complex task, encompassing comprehensive evaluation of various machine learning and deep learning algorithms. we experimented with support vector machines (SVM), random forest (RF) and long short-term memory (LSTM) algorithms incorporating various feature combinations involving Term Frequency – Inverse Document Frequency (TF-IDF) and Latent Dirichlet Allocation (LDA). The best performer of this comparative study was further evaluated against an LLM. The potential of large language models (LLMs), including their language understanding and prediction capabilities, is another key focus. We explore how these models could augment and advance mental health research, offering new perspectives and insights into the characterization of mental stress. Our findings show that the top model, an LSTM with TF-IDF and LDA (class weights assigned) outperformed the PaLM model with a coefficient of variation as low as 0.87% across all labels. Despite the PaLM model’s superior average performance, it exhibited higher variability among different labels.
IoT-Based Emergency Vehicle Detection Using YOLOv8 Syed Suhana, Boppuru Prathap Rudra, Kavish Narang, Ivin Anto Journal of Automation Mobile Robotics and Intelligent Systems, 2025 The Research focuses on the real-time identification of emergency vehicles using the YOLOv8 algorithm in the context of IoT. The aim is to develop an efficient and accurate emergency vehicle detection system to improve emergency service response times. The proposed system utilizes the YOLOv8 algorithm trained and tested with a dataset from a camera placed on a busy road. The results demonstrate that the system can detect emergency vehicles at a speed of 31 frames per second with a 95% accuracy rate. The system is implemented using a Raspberry Pi as an edge device, processing the live video stream from an IoT device equipped with a camera. Once an emergency vehicle is detected, an alert is sent to the emergency services for prompt action. The study highlights the potential of the YOLOv8 algorithm and IoT in creating effective and reliable emergency vehicle detection systems. The proposed solution is cost-effective, easy to implement, and adaptable to existing infrastructure. It has the capability to save lives and enhance emergency response by reducing response times. Future improvements can include the incorporation of more advanced machine learning algorithms and additional sensors to identify other emergency vehicles like ambulances and fire engines. The research emphasizes the potential of IoT and machine learning in developing innovative solutions for emergency services, particularly in the realm of intelligent transportation systems.
ONLINE CUSTOMERS’ PURCHASE INTENTIONS CLASSIFICATION USING HYBRID XGBOOST MACHINE LEARNING MODEL AND ITS SECURITY CONCERNS Attacks on Artificial Intelligence the New Facets of Cyber Ecospace, 2025
Comparative Study and Analysis of Prompting Techniques Using Gemini API Model and Reasoning Frameworks Rekha V, Prakash G L, Boppuru Rudra Prathap, Vandana Reddy Proceedings of 2025 International Conference on Computing for Sustainability and Intelligent Future Comp Sif 2025, 2025 Natural Language Processing (NLP) systems rely heavily on prompt engineering to enhance performance. This study evaluates various prompt techniques across different sizes of the Gemini API models (‘small’, ‘medium’, and ‘large’) using reasoning paradigms such as zero-shot, few-shot, Chain of Thought (CoT), and Tree of Thought (ToT). Metrics including accuracy, coherence, and latency were analyzed for prompts emphasizing specificity, brevity, example-driven accuracy, and creativity. Results indicate that the selection of reasoning paradigms and model sizes significantly impacts the performance, highlighting trade-offs between computational efficiency and quality. Recommendations for optimal configurations tailored to specific tasks are provided.
ATTACKS ON ARTIFICIAL INTELLIGENCE: The New Facets of Cyber Ecospace Attacks on Artificial Intelligence the New Facets of Cyber Ecospace, 2025
Ensemble-Based Phishing URL Detection: A Stacking Framework with Real-Time Deployment Surya S, Samarth Satish Airani, Anand Raju, Satvik Daruvuru, Boppuru Rudra Prathap 2025 International Conference on Computing and Communications Computingcon 2025, 2025 Phishing attacks deceive users through fraudulent websites to steal sensitive information, posing critical cybersecurity threats. This study develops a stacking ensemble framework combining Random Forest and XGBoost models for real-time phishing URL detection. Using a balanced dataset of 247,950 Uniform Resource Locators (URLs) with 41 extracted features including structural, domain-specific, and entropy-based measurements, the ensemble achieves 97.3% accuracy and 97.5% F1-score. Entropy-based features serve as primary discriminative indicators, whilst efficient batch processing enables realtime deployment through a browser extension. The framework demonstrates superior performance over individual classifiers through complementary strength integration and multi-modal feature fusion. The stacking ensemble approach improves phishing detection accuracy, but future work should address new attack techniques and scalability challenges.
Python Driven Keyword Analysis for SEO Optimization Aditya Pranav Kanara, Priya Kumari, Boppuru Rudra Prathap 10th International Conference on Advanced Computing and Communication Systems Icaccs 2024, 2024
A pragmatic study on heuristic algorithms for prediction and analysis of crime using social media data Journal of Advanced Research in Dynamical and Control Systems, 2019
