Harshal Dalvi

@djsce.ac.in

Assistant Professor, Dept. of Information Technology
SVKM's Dwarkadas J.Sanghvi College of Engineering, Mumbai

Harshal Dalvi


               Download Compact PDF  
https://researchid.co/harshal.dalvi

RESEARCH INTERESTS

Artificial Intelligence, Machine Learning, Software Engineering, Blockchain
24

Scopus Publications

222

Scholar Citations

7

Scholar h-index

5

Scholar i10-index

Scopus Publications

  • Lawbot: An Enhanced Legal Information Retrieval System using RAG
    Anvaya Solanki, Himanshu Main, Dhara Mehta, Devesh Kulkarni, Harshal Dalvi
    Conference Proceedings 2025 IEEE Silchar Subsection Conference IEEE Silcon 2025, 2025
    A large section of the public still remains legally illiterate. This lack of legal education is more common among low-income groups, marginalized communities, and the rural population. Furthermore, the absence of understandable legal information delays the judicial process which adds to the backlog of millions of pending cases. This paper presents Lawbot, a Retrieval-Augmented Generation (RAG)-based legal information retrieval system designed to improve public access to legal knowledge. Integrating the Kanoon API for authentic case retrieval and embedding-based semantic search, Lawbot provides modules for chat-based queries, document summarization, and real-time legal news and blogs. The system achieved high accuracy for user satisfaction in survey-based evaluations. Compared to existing Lawbot demonstrates superior contextual understanding and explainability through RAG-driven augmentation. The results highlight its potential as an accessible and scalable solution for enhancing legal literacy and democratizing legal information access.
  • Multiattribute Deep CNN-Based Approach for Detecting Medicinal Plants and Their Use for Skin Diseases
    Prachi Dalvi, Dhananjay R. Kalbande, Surendra Singh Rathod, Harshal Dalvi, Amey Agarwal
    IEEE Transactions on Artificial Intelligence, 2025
    Skin health is a critical concern for humans, especially in geographical areas where environmental conditions and lifestyle factors adversely affect their condition, leading to a prevalence of skin diseases. This issue is exacerbated in rural regions, like parts of India, where a notable dermatologist shortage exists, leading to overlooked skin diseases. In response, the use of medicinal plants for dermatological purposes has been a longstanding tradition. However, traditional plant identification often relies on a single attribute, such as leaves or flowers, which can be unreliable due to seasonal variations. This article introduces a novel approach for accurately identifying medicinal plants using a multiattribute deep convolutional neural network. This approach aims to bridge the gap in healthcare access by empowering individuals to recognize and utilize these plants effectively. Our objective is to develop a robust deep CNN model trained on a diverse dataset of images encompassing leaves, trunks, and seeds of medicinal plants associated with skin health. Our findings demonstrate that the model achieves high accuracy in plant identification, effectively addressing the limitations of single-attribute methods. This research not only contributes to the field of medicinal plant classification but also empowers individuals to make informed decisions about their skin health while preserving valuable traditional knowledge.
  • Explainable Models for the Detection of Incidents of Fake News and Hate Speech
    Vraj Desai, Ashray Gattani, Harshal Dalvi
    Text and Social Media Analytics for Fake News and Hate Speech Detection, 2024
    The current existence of social media and its involvement in our society has granted people the freedom to develop new ways to interact and exchange information on a large scale leading to the spread of fake news and hate speech. Recently, there have been several AI-based analytical attempts developed that deal with the perception of fake news phenomena which uses the identification of typical fake news patterns, options, and hate speech detection. These AI-based models are unbiased in the sense that their features are arbitrarily chosen from the pool of accessible features. Using explainable artificial intelligence (XAI) we can counter these issues by providing more transparent and more clear explanations for the output given by traditional models. XAI techniques can help to identify a strong interconnection between the keywords and fake news detection patterns, essential for research toward achieving transparent user–machine interaction. In this chapter, we will be analyzing different XAI techniques available viz. LIME, SHAP, and eli5. The selected set of features are passed to global and local explainable models such as SHAP and LIME to generate probabilities for the features selected and plot the probabilities in a graphical format. This method can help machine learning (ML) practitioners to better understand the rationale behind model decisions and labeling a news article as fake/hate speech or as a genuine one. This can further help ML practitioners finetune the necessary parameters to improve model performance and address the trade-off between model accuracy and transparency. Experimental analysis of fake news and hate speech datasets using logistic regression determines that our proposed explainable model achieves remarkable classification of fake or real, hate or not hate.
  • Employing Explainable AI to Optimise Domestic Energy for a Greener Society
    Vraj Desai, Ashray Gattani, Harshal Dalvi, Meera Narvekar
    Lecture Notes in Electrical Engineering, 2024
  • Early Detection and Diagnosis of Brain Related Diseases
    Karna Mehta, Preet Anam, Parshva Vyas, Harshal Dalvi
    Communications in Computer and Information Science, 2024
  • An Analytical Approach and Concept Mapping of Agricultural Issues Using Deep Learning Techniques
    Ashray Gattani, Pradhyuman Pandey, Harshal Dalvi, Neha Katre
    Studies in Computational Intelligence, 2024
  • Automated Pothole Detection using Transfer Learning
    Karan Thakkar, Saurav Shah, Bhavna Mulchandani, Neha Katre, Harshal Dalvi
    2024 IEEE 9th International Conference for Convergence in Technology I2ct 2024, 2024
    Road irregularities like potholes, which are common and persistent, provide a serious problem for both drivers and the government. In order to address this long-standing problem, the research study examines a novel strategy that combines transfer learning with ResNet (Residual Neural Network) methodologies. The main cause of pothole formation is unfavourable weather, which gradually erodes road surfaces and leaves voids behind. Potholes are created as a result of these voids expanding over time and are frequently hidden by flooding or rain during bad weather. Current government responses frequently suffer from delays and resource shortages and rely on manual inspections and reactive repair plans. The paper suggests developing real-time, inexpensive pothole detecting systems with a remarkable accuracy rate of 98.24 percent by utilising ResNet and transfer learning. These solutions seamlessly incorporate the locations of discovered potholes, along with geo-tagging tools that accurately identify each pothole’s coordinates and map them to Google Maps for public access and awareness. This proactive strategy enables authorities to quickly identify and prioritise maintenance activities, greatly improving road safety, lowering the cost of vehicle maintenance, and efficiently allocating resources. In the end, while being specifically created for use by government officials, this technology-driven strategy offers a significant improvement over current methods and stands to benefit the public by raising the calibre of transportation infrastructure.
  • Efficient Data Extraction from Handwritten Forms: A Structured Pipeline Solution
    Shivani Patel, Krisha Borana, Neha Katre, Vinaya Sawant, Harshal Dalvi
    2024 2nd International Conference on Artificial Intelligence Trends and Pattern Recognition Icaitpr 2024, 2024
    This paper presents an approach to systematically extract data from handwritten forms and convert them to structured data. Unlike traditional handwriting character recognition systems, proposed approach integrates techniques such as alpha blending for precise image alignment and deep learning-based models for enhanced text recognition accuracy. The pipeline begins with the automated detection and delineation of form fields (like Name, DOB, Occupation) followed by sophisticated pre-processing steps that include noise reduction and threshold management to ensure optimal image quality. The system then employs the proposed trained model to recognize handwritten text and match it to the extracted entities like key value pairs. The values are then appended to structured data. The proposed method has been evaluated, demonstrating accuracy and reliability in diverse real-world scenarios. This method would bypass the need to individually extract values and handwritten texts, saving time and effort.
  • Interpreting Accuracy: A Comprehensive Analysis of AI Models for Liver Disease Diagnosis
    Harshal Dalvi, Meera Narvekar, Prachi Dalvi, Kunaal Vadgama, Momin Kasmani, Krish Ranawat
    2024 Parul International Conference on Engineering and Technology Picet 2024, 2024
    It is critical to make a timely diagnosis of liver disease to avoid fatal health complications. The use of AI-based diagnosis systems can help in the early detection of such liver diseases based on health parameters. However, the use of such systems is not yet widely accepted by medical practitioners due to their complex nature and inability to comprehend the rationale behind the outcomes. In this research article, we attempted to identify the relationship between interpretability and accuracy in machine learning models applied for diagnosing liver disease using a variety of models, such as Logistic Regression, KNN, decision tree, XGBoost, and random forest. We performed a thorough evaluation of a dataset of liver patients and analyzed traditional performance metrics, such as accuracy, recall, precision, and F1 score, to provide a thorough insight into the diagnostic capabilities of each model. Moreover, we use SHAP (SHapley Additive exPlanations) values to decipher the underlying characteristics that propel the predictions of the models. The goal of this interpretability analysis is to clarify the decision-making process and provide important clinical practitioner insights. By carefully analyzing interpretability and accuracy, we hope to uncover possible trade-offs and offer insightful advice for using these models in actual healthcare environments. This work has immediate implications for clinical practice in addition to adding to the growing body of knowledge in medical diagnostics and machine learning. The information provided here aims to improve patient care and outcomes by strengthening the validity and efficacy of liver disease diagnosis.
  • An AI-Based Adaptive Assessment System for Effective Campus Placement Process Management
    Shivam Vora, Alankrit Arya, Chaitanya Kumbhar, Harshal Dalvi
    Aip Conference Proceedings, 2023
  • Early Diagnosis of Cataract and Diabetic Retinopathy for Rural India using a Cloud-Based Deep Learning Model
    Naitik Vora, Vighnesh Iyer, Harshal Dalvi
    Proceedings of 3rd International Conference on Advanced Computing Technologies and Applications Icacta 2023, 2023
  • XAI Meets Ophthalmology: An Explainable Approach to Cataract Detection Using VGG-19 and Grad-CAM
    Harsh Shah, Rutwik Patel, Sanchit Hegde, Harshal Dalvi
    2023 IEEE Pune Section International Conference Punecon 2023, 2023
  • Blockchain-based reliable supply chain management (SCM) for vaccine distribution and traceability using identity management approach
    Shaurya Magar, Manan Doshi, Shazia Talib, Harshal Dalvi
    Unleashing the Potentials of Blockchain Technology for Healthcare Industries, 2023
  • Decentralized Ride Hailing System using Blockchain and IPFS
    Om Naik, Nimai Patel, Sabir Ali Baba, Harshal Dalvi
    Ibssc 2022 IEEE Bombay Section Signature Conference, 2022
  • A Decentralized Approach for Evidence Management in Archaeology
    Rishabh Nevatia, Arya Patki, Onkar Bagwe, Harshal Dalvi, Neha Katre
    Proceedings of the 2022 9th International Conference on Computing for Sustainable Global Development Indiacom 2022, 2022
  • A Study of LIME and SHAP Model Explainers for Autonomous Disease Predictions
    Sannidhi Rao, Shikha Mehta, Shreya Kulkarni, Harshal Dalvi, Neha Katre, Meera Narvekar
    Ibssc 2022 IEEE Bombay Section Signature Conference, 2022
  • Serverless Computing and the Emergence of Function-as-a-Service
    Rishabh Patil, Tanveesh Singh Chaudhery, Muhammad Ali Qureshi, Vinaya Sawant, Harshal Dalvi
    2021 6th International Conference on Recent Trends on Electronics Information Communication and Technology Rteict 2021, 2021
  • Application of AI As Singing Trainer
    Vivek Vinze, Jainam Dhami, Darshit Desai, Harshal Dalvi, Purva Raut
    2021 7th IEEE International Conference on Advances in Computing Communication and Control Icac3 2021, 2021
  • Blockchain-based Universal Loyalty Platform
    Mausam Agrawal, Divya Amin, Harshal Dalvi, Riken Gala
    2019 6th IEEE International Conference on Advances in Computing Communication and Control Icac3 2019, 2019
  • Automated system for detecting distracted driver
    Vyom Unadkat, Parth Sayani, Harshal Kapadia, Prasham Shah, Harshal Dalvi
    2018 4th International Conference on Computing Communication and Automation Iccca 2018, 2018
  • Trustworthiness evaluation system in E-Commerce context
    Harshal D. Dalvi, Abhijit Joshi, Narendra Shekokar
    Proceedings 2nd International Conference on Computing Communication Control and Automation Iccubea 2016, 2017
  • Developing hedging strategies in option segment
    Sagar Sanghvi, Harsh Shah, Suryansh Haria, Abhijit R. Joshi, Harshal Dalvi
    Proceedings 2nd International Conference on Computing Communication Control and Automation Iccubea 2016, 2017
  • Querying the database with XQuery to optimize application performance
    11th Indiacom 4th International Conference on Computing for Sustainable Global Development Indiacom 2017, 2017
  • CompareKart
    Jay Parikh, Rahanik Vora, Niti Doshi, Akshay Waghela, Abhijit Joshi, Harshal Dalvi
    Procedia Computer Science, 2015

RECENT SCHOLAR PUBLICATIONS

  • Efficient Data Extraction from Handwritten Forms: A Structured Pipeline Solution
    S Patel, K Borana, N Katre, V Sawant, H Dalvi
    2024 2nd International Conference on Artificial Intelligence Trends and … , 2024
    2024
    Citations: 1
  • Multiattribute deep CNN-based approach for detecting medicinal plants and their use for skin diseases
    P Dalvi, DR Kalbande, SS Rathod, H Dalvi, A Agarwal
    IEEE Transactions on Artificial Intelligence 6 (3), 710-724 , 2024
    2024
    Citations: 18
  • An Analytical Approach and Concept Mapping of Agricultural Issues Using Deep Learning Techniques
    A Gattani, P Pandey, H Dalvi, N Katre
    Computational Intelligence in Internet of Agricultural Things, 147-169 , 2024
    2024
  • Explainable models for the detection of incidents of fake news and hate speech
    V Desai, A Gattani, H Dalvi
    Text and Social Media Analytics for Fake News and Hate Speech Detection, 114-136 , 2024
    2024
    Citations: 2
  • Early Detection and Diagnosis of Brain Related Diseases
    K Mehta, P Anam, P Vyas, H Dalvi
    International Conference on Intelligent Computing and Big Data Analytics, 72-88 , 2024
    2024
  • Mitigating Data Bias In Machine Learning: Enhancing Model Transparency Through Fairness-Aware Techniques.
    H Dalvi, M Narvekar, S Shah, P Donda, J Shah
    Frontiers in Health Informatics 13 (4) , 2024
    2024
  • Interpreting Accuracy: A Comprehensive Analysis of AI Models for Liver Disease Diagnosis
    H Dalvi, M Narvekar, P Dalvi, K Vadgama, M Kasmani, K Ranawat
    2024 Parul International Conference on Engineering and Technology (PICET), 1-6 , 2024
    2024
    Citations: 1
  • Automated Pothole Detection using Transfer Learning
    K Thakkar, S Shah, B Mulchandani, N Katre, H Dalvi
    2024 IEEE 9th International Conference for Convergence in Technology (I2CT), 1-8 , 2024
    2024
    Citations: 4
  • Privacy-Centric Approach in Leveraging Federated Learning for Improved Parkinson's Disease Diagnosis
    RA Joshi, SR Mangle, N Katre, H Dalvi
    Pioneering Smart Healthcare 5.0 with IoT, Federated Learning, and Cloud … , 2024
    2024
    Citations: 2
  • Xai meets ophthalmology: an explainable approach to cataract detection using vgg-19 and grad-cam
    H Shah, R Patel, S Hegde, H Dalvi
    2023 IEEE Pune Section International Conference (PuneCon), 1-8 , 2023
    2023
    Citations: 10
  • An AI-based adaptive assessment system for effective campus placement process management
    S Vora, A Arya, C Kumbhar, H Dalvi
    AIP Conference Proceedings 2916 (1), 020014 , 2023
    2023
    Citations: 2
  • Early diagnosis of cataract and diabetic retinopathy for rural india using a cloud-based deep learning model
    N Vora, V Iyer, H Dalvi
    2023 International Conference on Advanced Computing Technologies and … , 2023
    2023
    Citations: 3
  • Employing Explainable AI to Optimise Domestic Energy for a Greener Society
    V Desai, A Gattani, H Dalvi, M Narvekar
    International Conference on Entrepreneurship, Innovation, and Leadership … , 2023
    2023
  • VoteCoin: A Blockchain Perspective on E-Voting Infrastructure
    A Kotey, K Gupta, J Agarwal, H Dalvi, N Khatre
    International Journal of Scientific Research in Computer Science … , 2023
    2023
  • Blockchain-based reliable supply chain management (SCM) for vaccine distribution and traceability using identity management approach
    S Magar, M Doshi, S Talib, H Dalvi
    Unleashing the Potentials of Blockchain Technology for Healthcare Industries … , 2023
    2023
    Citations: 5
  • A study of LIME and SHAP model explainers for autonomous disease predictions
    S Rao, S Mehta, S Kulkarni, H Dalvi, N Katre, M Narvekar
    2022 ieee bombay section signature conference (ibssc), 1-6 , 2022
    2022
    Citations: 61
  • Decentralized Ride Hailing System using Blockchain and IPFS
    O Naik, N Patel, SA Baba, H Dalvi
    2022 IEEE Bombay Section Signature Conference (IBSSC), 1-5 , 2022
    2022
    Citations: 2
  • A Decentralized Approach for Evidence Management in Archaeology
    R Nevatia, A Patki, O Bagwe, H Dalvi, N Katre
    2022 9th International Conference on Computing for Sustainable Global … , 2022
    2022
    Citations: 1
  • MyRoomie: a roommate finding app
    V Tikoo, R Jain, P Sheth, H Dalvi
    International Journal for Research 10 (3), 1937-1943 , 2022
    2022
    Citations: 3
  • Application of AI as singing trainer
    V Vinze, J Dhami, D Desai, H Dalvi, P Raut
    2021 International Conference on Advances in Computing, Communication, and … , 2021
    2021
    Citations: 3

MOST CITED SCHOLAR PUBLICATIONS

  • A study of LIME and SHAP model explainers for autonomous disease predictions
    S Rao, S Mehta, S Kulkarni, H Dalvi, N Katre, M Narvekar
    2022 ieee bombay section signature conference (ibssc), 1-6 , 2022
    2022
    Citations: 61
  • Blockchain-based universal loyalty platform
    M Agrawal, D Amin, H Dalvi, R Gala
    2019 international conference on advances in computing, communication and … , 2019
    2019
    Citations: 39
  • Multiattribute deep CNN-based approach for detecting medicinal plants and their use for skin diseases
    P Dalvi, DR Kalbande, SS Rathod, H Dalvi, A Agarwal
    IEEE Transactions on Artificial Intelligence 6 (3), 710-724 , 2024
    2024
    Citations: 18
  • Automated system for detecting distracted driver
    V Unadkat, P Sayani, H Kapadia, P Shah, H Dalvi
    2018 4th International Conference on Computing Communication and Automation … , 2018
    2018
    Citations: 16
  • Xai meets ophthalmology: an explainable approach to cataract detection using vgg-19 and grad-cam
    H Shah, R Patel, S Hegde, H Dalvi
    2023 IEEE Pune Section International Conference (PuneCon), 1-8 , 2023
    2023
    Citations: 10
  • System to fight counterfeit drugs
    S Tendulkar, A Rodrigues, K Patel, H Dalvi
    Advanced Computing Technologies and Applications: Proceedings of 2nd … , 2020
    2020
    Citations: 9
  • Serverless computing and the emergence of function-as-a-service
    R Patil, TS Chaudhery, MA Qureshi, V Sawant, H Dalvi
    2021 International Conference on Recent Trends on Electronics, Information … , 2021
    2021
    Citations: 8
  • Advanced computing technologies and applications
    M Mhapsekar, P Mhapsekar, A Mhatre, V Sawant
    Springer , 2020
    2020
    Citations: 6
  • Blockchain-based reliable supply chain management (SCM) for vaccine distribution and traceability using identity management approach
    S Magar, M Doshi, S Talib, H Dalvi
    Unleashing the Potentials of Blockchain Technology for Healthcare Industries … , 2023
    2023
    Citations: 5
  • Blockchain-powered real estate system
    A Jain, B Chitroda, A Dixit, H Dalvi
    Advanced Computing Technologies and Applications: Proceedings of 2nd … , 2020
    2020
    Citations: 5
  • Advanced computing technologies and applications
    H Vasudevan, A Michalas, N Shekokar, M Narvekar
    Proc. 2nd Int. Conf. Adv. Comput. Technol. Appl.(ICACTA), 300 , 2020
    2020
    Citations: 5
  • Augmented Reality Books: An Immersive Approach to Learning
    S Mehta, P Jain, A Vora, A Joshi, H Dalvi
    8th International Conference on Computing, Communication, Control and … , 2017
    2017
    Citations: 5
  • Automated Pothole Detection using Transfer Learning
    K Thakkar, S Shah, B Mulchandani, N Katre, H Dalvi
    2024 IEEE 9th International Conference for Convergence in Technology (I2CT), 1-8 , 2024
    2024
    Citations: 4
  • Trustworthiness evaluation system in E-Commerce context
    HD Dalvi, A Joshi, N Shekokar
    2016 International Conference on Computing Communication Control and … , 2016
    2016
    Citations: 4
  • Early diagnosis of cataract and diabetic retinopathy for rural india using a cloud-based deep learning model
    N Vora, V Iyer, H Dalvi
    2023 International Conference on Advanced Computing Technologies and … , 2023
    2023
    Citations: 3
  • MyRoomie: a roommate finding app
    V Tikoo, R Jain, P Sheth, H Dalvi
    International Journal for Research 10 (3), 1937-1943 , 2022
    2022
    Citations: 3
  • Application of AI as singing trainer
    V Vinze, J Dhami, D Desai, H Dalvi, P Raut
    2021 International Conference on Advances in Computing, Communication, and … , 2021
    2021
    Citations: 3
  • Developing hedging strategies in Option segment
    S Sanghvi, H Shah, S Haria, AR Joshi, H Dalvi
    2016 International Conference on Computing Communication Control and … , 2016
    2016
    Citations: 3
  • Explainable models for the detection of incidents of fake news and hate speech
    V Desai, A Gattani, H Dalvi
    Text and Social Media Analytics for Fake News and Hate Speech Detection, 114-136 , 2024
    2024
    Citations: 2
  • Privacy-Centric Approach in Leveraging Federated Learning for Improved Parkinson's Disease Diagnosis
    RA Joshi, SR Mangle, N Katre, H Dalvi
    Pioneering Smart Healthcare 5.0 with IoT, Federated Learning, and Cloud … , 2024
    2024
    Citations: 2