Dr. KARTHIK KOVURI

@bvrit.ac.in

Dean, Academics and Professor, Computer Science and Engineering
B V Raju Institute of Technology

Dr. KARTHIK KOVURI


                     Download Compact PDF  
https://researchid.co/karthikkovuri
Dr. Karthik Kovuri is an Academician, Researcher and an Administrator. He has 18 years of teaching experience at various academic and administrative positions in Engineering Institutions. At present, he is working as Associate Dean for Academics, and Professor in the Department of Computer Science & Engineering at B.V. Raju Institute of Technology, Narsapur, Telangana.
He designs curriculum for UG and PG technological education. His areas of interest include Computer Networks, Architectures, Parallel Processing, Algorithms, Cloud Computing, Blockchain, IoT and Ubiquitous Computing. He has published 28 research papers in National & International Journals and Conferences.
He represented India in the MIC Global Summit at Seattle, USA. He is the recipient of Indian Academic Leader Award-2022, Jyeshta Acharya Award, Microsoft Azure Educator Grant Award and Outstanding Faculty Teaching Award. He has written a book on “Strategic perspectives of Cloud Computing” and has 5 Patents.

EDUCATION

B.Tech. in Computer Science and Engineering
M.Tech. in Computer Science and Engineering
Ph.D in Computer Science and Engineering

RESEARCH INTERESTS

Computer Networks, Architectures, Parallel Processing, Algorithms, Cloud Computing, Blockchain, IoT and Ubiquitous Computing.
20

Scopus Publications

Scopus Publications

  • Empowering E-commerce Decisions: Machine Learning Insights from Customer Review Sentiments
    Karthik Kovuri, K. Reddy Madhavi, K. Meenu Yadav, E. Vivek, K. Sohith Kumar Raju, K. Sree Preethi, Mohammad Gouse Galety
    Lecture Notes in Networks and Systems, 2026
  • Vulnerabilities in IoT device clusters: Analysing security weaknesses and attack impacts
    Prabhat Das, Atrayee Banerjee, Partha Mandal, Balaram Chakraborty, Arya Singh, Kathik Kovuri, Sajal Saha, Bhagirath Tallapragada, Varchaleswari Ganugapati
    Data Driven AI A Multidisciplinary Approach Techniques Applications and Insights from Multiple Domains, 2026
    Although the Internet of Things (IoT) has transformed connectivity in smart homes, industrial systems, and urban infrastructure, it is still extremely susceptible to changing security risks. This paper simulates Deauthentication, Man-in-the-Middle, Evil Twin, and DDoS attacks against WPA2/WPA3 protocols in order to examine critical vulnerabilities within IoT clusters. Using Raspberry Pi boards and ESP32 microcontrollers, we have set up IoT device clusters to gather network traffic in both hostile and benign environments. To create a labelled dataset for supervised classification, packet flows are classified as benign or malicious using unsupervised real-time clustering using the BIRCH algorithm. After evaluating several machine learning classifiers, XGBoost showed the best detection accuracy and resilience in differentiating between legitimate and attack traffic. Persistent IoT vulnerabilities, such as insufficient authentication methods, constrained processing power, and an excessive dependence on Wi-Fi protocols, are revealed by our analysis. In order to secure resource-constrained IoT clusters and increase the resilience of linked systems, the suggested clustering-guided labelling in conjunction with supervised learning provides a scalable, real-time solution.
  • SCALABILITY AND EFFICIENCY OF CLUSTERING ALGORITHMS FOR LARGE-SCALE IoT DATA: A COMPARATIVE ANALYSIS
    Journal of Theoretical and Applied Information Technology, 2025
  • 'Time and Energy Trade-offs for Mobile Edge Computing: A Comparative Study of Task Offloading Strategies'
    Md. Mainul Hoque, Karthik Kovuri
    2025 1st International Conference on Aiml Applications for Engineering and Technology Icaet 2025, 2025
    Mobile Edge Computing (MEC) has emerged as a promising paradigm to enable compute-intensive and latency-critical applications on resource-constrained mobile devices by offloading computation tasks to nearby MEC servers. However, offloading also consumes energy due to wireless transmission. Therefore, making optimal offloading decisions to achieve both time and energy efficiency is an important research challenge. This paper provides a comprehensive literature review of various computation offloading strategies proposed for MEC systems concerning joint optimization of execution time and energy consumption. The offloading strategies are classified into three categories - static, dynamic with partial knowledge, and dynamic with full knowledge. Under each class, various mathematical modeling approaches and optimization objectives are discussed. Open research issues and future directions for time and energy-efficient offloading in MEC are also highlighted.
  • Enhanced Clustering Framework for Unveiling Hidden Patterns in IoT-Based Structured and Unstructured Data
    Prabhat Das, Karthik Kovuri, Sajal Saha
    Proceedings of 2025 3rd International Conference on Intelligent Systems Advanced Computing and Communication Isacc 2025, 2025
    Clustering algorithms play a pivotal role in uncovering hidden patterns and relationships within diverse datasets. This study discusses about an enhanced real-time clustering approach based on the Birch methodology, optimized for analyzing both structured and unstructured data. Structured data, consisting of IoT sensor readings, and unstructured image data, captured using Raspberry Pi devices deployed in various locations, were analyzed to identify meaningful patterns. For structured data, the algorithm effectively clustered environmental parameters, revealing correlations and anomalies. For unstructured data, high-dimensional image features extracted using a pre-trained VGG16 model were clustered, uncovering patterns in texture, color, and structure. Experimental results demonstrate the algorithm’s scalability and efficiency, achieving high-quality clusters with average Silhouette Scores of 0.87 and 0.81 for structured and unstructured datasets, respectively. These findings highlight the algorithm’s adaptability and its potential for real-time analytics in diverse applications such as environmental monitoring and image-based decision-making systems.
  • Innovative Predictive Framework for King County Housing Prices: A Comparative Study of Machine Learning and Ensemble Approaches
    Manjot Kaur Sidhu, Sugandh Singh, Shilpa Jain, Karthik Kovuri, Anshu Tomar
    2025 7th International Symposium on Advanced Electrical and Communication Technologies Isaect 2025, 2025
    Effective prediction of the residential property values is crucial to the stakeholders in the fluctuating real estate markets. This paper creates a powerful machine learning pipeline to forecast house prices on the basis of the King County data that include 21,613 records. Preprocessing of data consisted of the removal of outliers by interquartile range method and feature engineering, which added a composite attribute, total-sqft. Feature selection using correlation and principal component analysis were used to reduce dimensionality and retain 95 percent variance. Linear Regression, Decision tree, Random Forest, XGBoost, LightGBM and a Stacking Ensemble with 5fold cross-validation were tested on each using grid search. The best performance was the Stacking Ensemble, and it has the highest R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> of 0.92 and the lowest MSE followed by LightGBM (R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> = 0.91). Model efficacy and contributions made by features were clarified through the use of visualization such as correlation heatmap, MSE / R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> bar plot, scatter plot, residual histogram, and feature importance chart. Important results emphasize the effectiveness of the ensemble approach in the preservation of nonlinear trends and the importance of the manipulated properties such as the feature of the total in the form of the total_sqft.
  • Sustainable Business Models Powered by AI and Blockchain
    Md. Arshad Ur Rahman, Praveen Singh, Karthik Kovuri, Prateek Negi, Kirti Udayai, Vasim Ahmad
    13th International Conference on Intelligent Embedded Microelectronics Communication and Optical Networks Iemecon 2025, 2025
    The purpose of sustainable business models is to make the same money but do less damage. In order to create an effective sustainable business model; you need to have reliable data, know what is happening operationally and understand how to comply with regulatory requirements on a consistent basis. Many of these issues may be resolved through blockchain technology and AI. AI can help facilitate data driven decision making with the use of predictive analytics to optimize resource utilization and track activity in real time. Blockchain can assist with ensuring that data is transparent, traceable and stored securely which will allow for responsible sourcing, carbon tracking and ESG reporting. This paper reviews the possibilities, barriers, and future potential of using AI and/or blockchain individually or collectively to support sustainability in core business areas (such as supply chain management, energy efficiency, compliance reporting, and stakeholder communication). This paper provides a number of case studies, examples and analysis of current uses to provide insight into the potential of AI and/or blockchain to support the creation of sustainable business models.
  • Data Privacy in Voice Search Advertising: A Blockchain-Based Consent Model
    Mohd Arif Hussain, Vivek Verma, Karthik Kovuri, Aishwarya Pratap, Vivek Shrivastava, Prasenjeet Samanta
    2025 International Conference on Future Technologies Icft 2025, 2025
  • The Role of Robotic Process Automation in Computerized Accounting Workflows
    Karthik Kovuri, Vaishali Gaur, Amanjot Kaur Hira, Md Sikandar Azam, Ruchitra Kaparwan, Vasim Ahmad
    2025 IEEE International Conference on Advanced Computing Technologies Icact 2025, 2025
    Robotic Process Automation (RPA) is redefining accounting processes by automating time-consuming and redundant processes, in the process, driving enhanced efficiency, accuracy, and compliance. The study elaborates on how computerized accounting processes, such as payroll processing automation, automatic generation of financial statements, automatic calculation of taxes, and assistance in budget, are being redefined by RPA. Implementation hindrances to RPA, including legacy system integration, resistance to change, and data protection, have also been elaborated. The study elaborates on emerging trends such as fusion of RPA and Artificial Intelligence (AI) and Machine Learning (ML) and emergence of hyper-automation and increased application of predictive analytics in accounting. Practical insights on how to ensure optimal benefits through RPA and overcome implementation barriers have been given through case studies. The paper concludes by briefly mentioning changed roles of accounting professionals in an age of automation and factors taken into account in calculating the return on investment (ROI) in RPA-enabled accounting.
  • Meta-Learning Enhanced Self-Supervised Methods for Crop Yield Estimation Using UAV Imagery in Precision Agriculture
    Karthik Kovuri, V. Vishnu Vandana Devi, K. Ayyappa Swamy, K. Lavanya, J. Avanija, Sam Goundar
    Lecture Notes in Networks and Systems, 2025
  • Privacy-Preserving Healthcare Data in IoMT
    Karthik Kovuri, Gunipati Kanishka, Pattabhi Mary Jyosthna, A. Jyothi Babu, Nagendar Yamsani, R. Vanithamani
    Lecture Notes in Networks and Systems, 2025
  • Identification and Correction of Image Similarity using autoencoder
    Karthik Kovuri, Nampally Vijay Kumar, Awatef Balobaid, A. V. Sriharsha, Bura Vijay Kumar, K. Reddy Madhavi, A. Reni
    Lecture Notes in Networks and Systems, 2025
  • Deep neural networks for early diagnosis of neurodegenerative diseases
    K. Suneetha, Karthik Kovuri, Chengamma Chitteti, J. Avanija, K. Reddy Madhavi, Naresh Tangadu
    Deep Learning Approaches for Early Diagnosis of Neurodegenerative Diseases, 2024
  • Issues, Opportunities, and Limitations on the Convergence of Cybersecurity and Cloud Computing
    Karthik Kovuri, K. Reddy Madhavi, J. Avanija, Viswaksena Balaji, Harshavelu Irrigisetty, Chinthapatla Pranay Varna
    Convergence of Cybersecurity and Cloud Computing, 2024
  • Adaptable Fog Computing Framework for Healthcare 4.0
    Karthik Kovuri, Katha Chandrashekhar, A. V. Sriharsha, Byraboina Siddardha, A. Hitesh Reddy
    Lecture Notes in Networks and Systems, 2023
  • Performance Analysis of Interconnection Networks Using A Novel Metric
    Karthik Kovuri, Sudarson Jena, T Venu Gopal
    Ecs Transactions, 2022
  • Brain Stroke Prediction Using Deep Learning: A CNN Approach
    Madhavi K. Reddy, Karthik Kovuri, J Avanija, M Sakthivel, Shivaprasad Kaleru
    4th International Conference on Inventive Research in Computing Applications Icirca 2022 Proceedings, 2022
  • Evaluation and comparison of hypercube interconnection networks performance
    International Journal of Advanced Science and Technology, 2020
  • Performance index - A metric to analyze and evaluate performance of hypercube interconnection networks
    International Journal of Recent Technology and Engineering, 2018
  • Notice of Removal: Comparison of properties affecting the performance of Hypercube Interconnection networks
    K. Karthik, Sudarson Jena, T. Venu Gopal
    International Conference on Electrical Electronics Signals Communication and Optimization Eesco 2015, 2015

Publications

Journal Publications: 18
Conference Publications: 8