Sujit Nishikant Deshpande

@moderncoe.edu.in

Assistant Professor, Department of Computer Engineering
PES Modern College of Engineering Pune

Sujit Nishikant Deshpande


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

EDUCATION

Ph.D. in Computer Science and Engineering from Visvesvaraya Technological University, Belagavi.

RESEARCH, TEACHING, or OTHER INTERESTS

Computer Engineering, Computational Theory and Mathematics, Hardware and Architecture
6

Scopus Publications

Scopus Publications

  • Enhanced credit card fraud detection using boosting, stacking and feature importance analysis
    D.R. Patil, T.M. Pattewar, K.S. Kumavat, S.N. Deshpande, T.S. Shinde
    Journal of Computational Technologies, 2026
    Due to increasing complexity and frequency of credit card fraud, there is a critical need for highly accurate and efficient detection systems. This study proposes an enhanced fraud detection framework that combines ensemble learning with feature importance techniques to improve performance. It uses six powerful boosting algorithms — AdaBoost, XGBoost, GBM, LightGBM, CatBoost, and LogitBoost — as base models, which are then merged using a stacked ensemble method to boost prediction accuracy. To ensure model efficiency and interpretability, feature selection techniques such as recursive feature elimination, tree-based importance, mutual information classification, and ANOVA F-test are applied, with the ANOVA method prioritized in the final model. When evaluated on a benchmark dataset, the proposed system achieved exceptional results: accuracy, precision, recall, and F-measure of 99.97 %. This demonstrates the effectiveness of stacked ensembles in combining the strengths of individual models while minimizing errors. The feature selection process also improves computational efficiency by focusing on the most relevant features.
  • Two-Phase Video Encoding for Disaster Management Video Transmission
    Sushant M. Mangasuli, Arundhati V. Nelli, Ramesh Medar, Ranjana Battur, Sujit N. Deshpande
    Engineering Technology and Applied Science Research, 2025
    In disaster management scenarios, efficient video transmission poses significant challenges due to the limitations of existing encoding techniques. This study addresses these challenges by introducing a novel Two-Phase Video Encoding (TPVE) method tailored for disaster management video transmission. TPVE employs a two-phase design, reducing video bits through low-rank approximation in the initial phase, followed by additional compression using an optimized Huffman encoding in the second phase. The introduction of a parallel version of Huffman encoding with reduced coding length speeds up the encoding process, ensuring reduced Bit Error Rates (BERs) and higher coding efficiency. TPVE is particularly effective in Mobile Ad-Hoc Network (MANET) environments with Orthogonal Frequency-Division Multiplexing (OFDM) frames surrounded by noise and interference, and it outperforms Existing Video Encoding (EVE). Results reveal TPVE's superior BER and encoding efficiency, demonstrating a remarkable 65.06% BER improvement for a MANET Edge Device (MED) antenna size of 32 and an outstanding 88.08% improvement for a size of 64. These findings establish TPVE as a transformative solution, significantly enhancing reliability, reducing storage requirements, optimizing bandwidth utilization, and ensuring superior encoding efficiency in disaster management video transmission.
  • Advancing dental implant classification through YOLO-based deep learning models
    Ashwini D. Khairkar, Sonali Kadam, Pankaj Kadam, Sujit Deshpande
    International Journal of Information Technology Singapore, 2025
  • VitaChat - An Enhanced Healthcare Assisting Chatbot
    Rashmi K. Dixit, Rohan P. Ankam, Sushma Gunjal, Sujit N. Deshpande, Nandini S. Gaddam
    2025 International Conference on Sensors and Related Networks Sennet 2025 Special Focus on Digital Healthcare 64220, 2025
    This study presents VitaChat, a sophisticated conversational AI framework intended to improve patients' and healthcare providers' access to trustworthy antibiotic-related information. The two main stages of the suggested system's operation are the processing and embedding of reliable medical texts and the retrieval and production of contextually appropriate responses. The first stage encodes information from three antibiotic reference books using the LangChain framework and a Sentence Transformer model. The embeddings are then stored in a ChromaDB vector store for effective similarity-based retrieval. In order to ensure transparency and reliability, the second phase uses a query-driven retrieval mechanism to provide precise, source-grounded responses with citations. Across a range of user skill levels, this integrated approach provides easy access to vital antibiotic information, such as indications, side effects, and usage. VitaChat's superior performance in response accuracy and usability is demonstrated by rigorous evaluation across a variety of query types and user scenarios, solidifying its position as a reliable tool for bridging the gap between accessible healthcare communication and complex medical knowledge. The suggested framework emphasises how machine learning and natural language processing can help healthcare settings make well-informed decisions.
  • A novel scheduling algorithm development and analysis for heterogeneous IoT protocol control system to achieve SCADA optimization: a next generation post covid solution
    Sujit N. Deshpande, Rashmi M. Jogdand
    International Journal of Information Technology Singapore, 2023
  • Industrial Automation by Development of Novel Scheduling Algorithm for Industrial IoT: IIoT Re-birth Out of Covid-19 Scenario
    Sujit N. Deshpande, Rashmi M. Jogdand
    Smart Innovation Systems and Technologies, 2022