Sudipta Hazra
Sudipta Hazra has a total of 28 co-authors. Here are the top co-authors by number of shared publications:
Siddhartha Chatterjee
Department of Computer Science and Engineering, College of Engineering and Management- Kolaghat, Kolaghat, West Bengal
2 shared publications- Amit Kumar Jha
Asansol Engineering College
1 shared publication - Ankita Mandal
Institute of Engineering and Management, Kolkata University of Engineering and Management
1 shared publication Anudeepa Gon
Assistant Professor, Computational Sciences, Brainware University (UGC approved), Barasat, India
1 shared publicationAnup Kumar Ghosh
Assistant Professor, Computer Science and Engineering, NSHM Knowledge Campus, Durgapur, India
1 shared publication- Anurag Sinha
Research Scholar, Department of Information Technology Guru Ghasidas Vishwavidyalaya Bilaspur Chhattisgarh India
1 shared publication
Latest Publications
- 1 April 2026
- Published by Springer Nature
- p. 207-218
A Survey on Deep Learning Based Plant Growth Prediction
- 1 April 2026
- Published by Springer Nature
- p. 207-218
A Survey on Deep Learning Based Plant Growth Prediction
- 29 July 2025
- Published by Springer Nature
- p. 173-182
Application of Artificial Intelligence in Supply Chain Management in Industry 4.0
- 29 July 2025
- Published by Springer Nature
- p. 173-182
Application of Artificial Intelligence in Supply Chain Management in Industry 4.0
- Open Access
- 22 May 2025
- Published by Institute of Central Computation and Knowledge Inc. in ICCK Transactions on Machine Intelligence
- Vol. 1 (1) , 17-28
IoT-Integrated Reinforcement Learning-Based Mine Detection System for Military and Humanitarian Applications
This research proposes an advanced system for landmine detection combining the internet of things and reinforcement learning, which seeks to resolve issues in conventional methods that misidentify more than 30% of detections, have slow reaction times, and are not suited for different environments. Others like metallic detectors and sniffer dogs also pose greater danger for wrong threat identification, more so due to sluggish attempts. The system proposed in this study is novel in that it customizes metal detection by integrating a sensor into military boots, thus permitting constant scanning without the use of hands. A meta-learning-based Machine Learning model improves detection accuracy. It was found that reward-driven reinforcement learning regulation improves mine detection accuracy, increases the analysis attempts in each evaluation phase, and alters the strategic settings. The range of analysis conducted during this study validates the argument in question but this reworking of the system does not fully resolve the issues. The innovation is having that with proper situational awareness this model enables real-time implementation of IoT devices. This adaptable system is not only advantageous for military endeavors but can also be useful for demining activities. More robust multisensory capabilities are essential to facilitate effective and safe landmine inspection all over the globe, so follow-up studies should concentrate on field trials with accompanying iterative improvements.
- 22 May 2025
- Published by Institute of Central Computation and Knowledge Inc. in ICCK Transactions on Machine Intelligence
- Vol. 1 (1) , 17-28
Open AccessIoT-Integrated Reinforcement Learning-Based Mine Detection System for Military and Humanitarian Applications
This research proposes an advanced system for landmine detection combining the internet of things and reinforcement learning, which seeks to resolve issues in conventional methods that misidentify more than 30% of detections, have slow reaction times, and are not suited for different environments. Others like metallic detectors and sniffer dogs also pose greater danger for wrong threat identification, more so due to sluggish attempts. The system proposed in this study is novel in that it customizes metal detection by integrating a sensor into military boots, thus permitting constant scanning without the use of hands. A meta-learning-based Machine Learning model improves detection accuracy. It was found that reward-driven reinforcement learning regulation improves mine detection accuracy, increases the analysis attempts in each evaluation phase, and alters the strategic settings. The range of analysis conducted during this study validates the argument in question but this reworking of the system does not fully resolve the issues. The innovation is having that with proper situational awareness this model enables real-time implementation of IoT devices. This adaptable system is not only advantageous for military endeavors but can also be useful for demining activities. More robust multisensory capabilities are essential to facilitate effective and safe landmine inspection all over the globe, so follow-up studies should concentrate on field trials with accompanying iterative improvements.
- Open Access
- 12 February 2025
- Published by F1000 Research Ltd in F1000Research
- Vol. 14, 193
Analyzing why AI struggles with drawing human hands with CLIP
Background Artificial Intelligence (AI) has made significant strides in various domains, but generating realistic human hands remains a challenge. This study explores the limitations of AI in capturing the fine details and proportions of hands, using Contrastive Language Image Pretraining (CLIP) as a case study. Methods Our analysis reveals that CLIP struggles to accurately represent hands due to inadequate training data, anatomical complexities, and practical challenges. We conducted a series of tests and analyses to identify the primary causes of CLIP’s difficulties. Results Our results show that CLIP’s struggles stem from data biases and insufficient anatomical representation in training datasets. Specifically, we found distorted finger relationships, inaccurate proportions, and deviations from expected hand geometry. Conclusion This study aims to provide a comprehensive examination of the current limitations and propose possible directions for future research. By leveraging CLIP for evaluation, control algorithms for structure enforcement, DALL-E for generation, AR for gesture tracking, and 3D modeling for anatomical accuracy, we can overcome the challenges of generating realistic human hands and advance AI’s capabilities in artistic creativity
- 12 February 2025
- Published by F1000 Research Ltd in F1000Research
- Vol. 14, 193
Open AccessAnalyzing why AI struggles with drawing human hands with CLIP
Background Artificial Intelligence (AI) has made significant strides in various domains, but generating realistic human hands remains a challenge. This study explores the limitations of AI in capturing the fine details and proportions of hands, using Contrastive Language Image Pretraining (CLIP) as a case study. Methods Our analysis reveals that CLIP struggles to accurately represent hands due to inadequate training data, anatomical complexities, and practical challenges. We conducted a series of tests and analyses to identify the primary causes of CLIP’s difficulties. Results Our results show that CLIP’s struggles stem from data biases and insufficient anatomical representation in training datasets. Specifically, we found distorted finger relationships, inaccurate proportions, and deviations from expected hand geometry. Conclusion This study aims to provide a comprehensive examination of the current limitations and propose possible directions for future research. By leveraging CLIP for evaluation, control algorithms for structure enforcement, DALL-E for generation, AR for gesture tracking, and 3D modeling for anatomical accuracy, we can overcome the challenges of generating realistic human hands and advance AI’s capabilities in artistic creativity
- 10 December 2024
- Published by Springer Nature
- p. 257-285
Review on Social and Ethical Concerns of Generative AI and IoT
- 10 December 2024
- Published by Springer Nature
- p. 257-285
Review on Social and Ethical Concerns of Generative AI and IoT
- Open Access
- 31 July 2024
- Published by Adroid Connectz Private Limited in International Journal on Computational Modelling Applications
- Vol. 1 (1) , 1-19
A Study on Anaphylactic-shock Forecasting System with Knowledge-Based Intelligent Architecture
A serious and potentially fatal allergic reaction Anaphylaxis requires prompt diagnosis and treatment. Accurate prediction of anaphylaxis is vital for preventing severe outcomes and accompanying medical management. This study investigates the creation and application of a hybrid knowledge-based system architecture to predict anaphylaxis. By combining rule-based systems with machine learning models, the goal is to integrate expert medical knowledge with data-driven insights, resulting in a robust, interpretable, and clinically useful predictive tool. Here the development of a hybrid Knowledge-Based Intelligent System Architecture is discussed. Also, this approach addresses a comprehensive study of the KBS-ML system’s capacity against traditional prediction methods, providing healthcare providers with timely and accurate identification of anaphylaxis risk, ultimately improving patient outcomes
- 31 July 2024
- Published by Adroid Connectz Private Limited in International Journal on Computational Modelling Applications
- Vol. 1 (1) , 1-19
Open AccessA Study on Anaphylactic-shock Forecasting System with Knowledge-Based Intelligent Architecture
A serious and potentially fatal allergic reaction Anaphylaxis requires prompt diagnosis and treatment. Accurate prediction of anaphylaxis is vital for preventing severe outcomes and accompanying medical management. This study investigates the creation and application of a hybrid knowledge-based system architecture to predict anaphylaxis. By combining rule-based systems with machine learning models, the goal is to integrate expert medical knowledge with data-driven insights, resulting in a robust, interpretable, and clinically useful predictive tool. Here the development of a hybrid Knowledge-Based Intelligent System Architecture is discussed. Also, this approach addresses a comprehensive study of the KBS-ML system’s capacity against traditional prediction methods, providing healthcare providers with timely and accurate identification of anaphylaxis risk, ultimately improving patient outcomes
- 16 May 2024
- Published by Springer Nature
- p. 205-214
Forecasting of Rainfall in Subdivisions of India Using Machine Learning
- 16 May 2024
- Published by Springer Nature
- p. 205-214
Forecasting of Rainfall in Subdivisions of India Using Machine Learning
- 25 July 2023
- Published by Springer Nature
- p. 539-549
An Analysis of Duckworth-Lewis-Stern Method in the Context of Interrupted Limited over Cricket Matches
- 25 July 2023
- Published by Springer Nature
- p. 539-549
An Analysis of Duckworth-Lewis-Stern Method in the Context of Interrupted Limited over Cricket Matches
- 5 May 2023
- Published by IGI Global Scientific Publishing
- p. 166-188
Application of Machine Learning Algorithms for Automatic Detection of Risk in Heart Disease
The main cause of death worldwide is heart disease, emphasizing the need for accurate risk prediction models to recognise those who are at high risk. In this study, we propose an automated approach using artificial intelligence to forecast the likelihood of developing heart disease. The dataset consists of various clinical and demographic features, including blood pressure, cholesterol levels, age, gender, and exercise habits. We evaluate the performance of numerous machine learning algorithms, such as neural networks, logistic regression, support vector machines, and random forests, in predicting the likelihood of heart disease. Our results demonstrate that automated learning techniques can effectively determine the likelihood of cardiac disease with high accuracy, precision, and recall. Furthermore, we conduct analysis feature importance to the risk prediction model can determine which factors have the greatest impact. The automated risk prediction system can provide early detection and intervention strategies for individuals at high risk, enabling proactive healthcare management and reducing the burden of heart disease. This research showcases the potential of machine learning algorithms in improving heart disease risk assessment and guiding personalized preventive measures. Machine learning which is employed in worldwide in different industries. In the healthcare industry, there are no exceptions. Determining whether or not there will be heart problems, abnormalities and other disorders can be highly dependent on machine learning. For the purpose of predicting probable heart conditions in humans, we are creating machine learning algorithms. In our work, We contrast the effectiveness of several classifiers, which includs Naive Bayes, Decision Tree, Logistic Regression, Random Forest, SVM. Finally, we evaluate the effectiveness of the suggested classifiers, including the more accurate Ada-boost and XG-boost. Early detection of heart disease in high-risk persons is essential for assisting them in deciding whether to alter their lifestyle, which reduces consequences. Medical data's hidden patterns may be exploited to diagnose health issues.
- 5 May 2023
- Published by IGI Global Scientific Publishing
- p. 166-188
Application of Machine Learning Algorithms for Automatic Detection of Risk in Heart Disease
The main cause of death worldwide is heart disease, emphasizing the need for accurate risk prediction models to recognise those who are at high risk. In this study, we propose an automated approach using artificial intelligence to forecast the likelihood of developing heart disease. The dataset consists of various clinical and demographic features, including blood pressure, cholesterol levels, age, gender, and exercise habits. We evaluate the performance of numerous machine learning algorithms, such as neural networks, logistic regression, support vector machines, and random forests, in predicting the likelihood of heart disease. Our results demonstrate that automated learning techniques can effectively determine the likelihood of cardiac disease with high accuracy, precision, and recall. Furthermore, we conduct analysis feature importance to the risk prediction model can determine which factors have the greatest impact. The automated risk prediction system can provide early detection and intervention strategies for individuals at high risk, enabling proactive healthcare management and reducing the burden of heart disease. This research showcases the potential of machine learning algorithms in improving heart disease risk assessment and guiding personalized preventive measures. Machine learning which is employed in worldwide in different industries. In the healthcare industry, there are no exceptions. Determining whether or not there will be heart problems, abnormalities and other disorders can be highly dependent on machine learning. For the purpose of predicting probable heart conditions in humans, we are creating machine learning algorithms. In our work, We contrast the effectiveness of several classifiers, which includs Naive Bayes, Decision Tree, Logistic Regression, Random Forest, SVM. Finally, we evaluate the effectiveness of the suggested classifiers, including the more accurate Ada-boost and XG-boost. Early detection of heart disease in high-risk persons is essential for assisting them in deciding whether to alter their lifestyle, which reduces consequences. Medical data's hidden patterns may be exploited to diagnose health issues.