A privacy preserving synthetic learner dataset for learning analytics in technology enhanced higher education Sanjay Agal Scientific Reports, 2026 The proliferation of technology-enhanced learning in higher education generates unprecedented student data, creating opportunities for learning analytics while raising critical privacy concerns. Current data sharing practices are severely constrained by privacy regulations and ethical considerations, impeding collaborative research and methodological advancement. This study addresses this fundamental tension by introducing SynEdu-HEDL, a comprehensive privacy-preserving synthetic dataset specifically designed for learning analytics in higher education.Developed through an innovative five-phase methodological framework integrating conditional tabular generative adversarial networks, temporal sequence generators, and differential privacy mechanisms, the dataset captures authentic educational patterns while ensuring robust privacy protection. SynEdu-HEDL comprises 20,000 synthetic student records encompassing 85 features across demographic characteristics, temporal learning interactions, engagement patterns, and academic performance metrics.A comprehensive three-dimensional validation framework evaluated SynEdu-HEDL across privacy protection, statistical fidelity, and analytical utility. Results demonstrate that the dataset provides strong privacy guarantees (membership inference AUC-ROC=0.512, statistically indistinguishable from random guessing), preserves essential statistical properties (average Wasserstein distance=0.043, correlation matrix similarity=94.1%), and supports diverse learning analytics tasks with models achieving performance within 1-5% of those trained on original data. Notably, transfer learning experiments show 24.4% performance improvement with only 10% real data, demonstrating practical value for resource-constrained settings.SynEdu-HEDL advances educational data science by providing a practical solution to data sharing barriers, supporting reproducible research, and establishing methodological standards for synthetic educational data validation. SynEdu-HEDL is openly available at https://github.com/drsanjayagal/SynEdu-HEDL and has been archived in Zenodo with the permanent DOI https://doi.org/10.5281/zenodo.18884938 . The repository includes comprehensive documentation, fostering community engagement and accelerating progress in privacy-preserving learning analytics.
A machine learning approach to risk based asset allocation in portfolio optimization Sanjay Agal, Krishna Raulji, Niyati Dhirubhai Odedra Scientific Reports, 2025 This paper introduces a novel machine learning framework for dynamic risk-based asset allocation that addresses fundamental limitations in traditional portfolio optimization methods. The proposed architecture integrates Long Short-Term Memory networks for volatility forecasting with differentiable risk budgeting layers and regime-switching mechanisms, enabling end-to-end training of portfolio weights under adaptive risk constraints. Unlike conventional approaches that rely on static risk budgets and historical covariance estimates, our methodology dynamically adjusts risk targets based on real-time market indicators, including volatility expectations, credit spreads, and yield curve dynamics. The framework achieves three primary research objectives: first, it demonstrates superior risk-adjusted performance with a Sharpe ratio of 1.38 during the out-of-sample period (2017-2022), representing a 55% improvement over traditional risk parity strategies and a 23% enhancement over contemporary deep learning approaches. Second, the architecture maintains computational efficiency through sparse attention mechanisms, scaling linearly with asset count while processing 50-asset portfolios in under 25 milliseconds. Third, the model preserves interpretability via SHAP-based risk attribution, providing transparent insights into allocation decisions across different market regimes. Empirical results reveal particularly strong performance during volatile market conditions, with maximum drawdowns reduced by 41% during stress periods compared to conventional methods. The framework's proactive risk management capabilities were evidenced during the COVID-19 crisis, where it began reducing equity exposure two weeks before the market trough, demonstrating genuine predictive ability rather than reactive adjustment. Robustness checks confirm performance persistence under varying transaction costs, rebalancing frequencies, and alternative risk measures. These findings establish a new paradigm for portfolio optimization that successfully bridges theoretical finance with practical implementation. The framework's ability to navigate complex market environments while maintaining computational efficiency and interpretability suggests readiness for widespread institutional adoption. This research contributes to the evolving literature on differentiable finance while providing portfolio managers with a robust tool for constructing adaptive, risk-aware investment strategies.
Spatiotemporal Graph Networks for Relational Reasoning in Campus Infrastructure Management Sanjay Agal, Krishna Raulji, Nikunj Bhavsar, Pooja Bhatt International Journal of Advanced Computer Science and Applications, 2025 The efficient management of campus infrastructure presents a complex spatiotemporal forecasting challenge characterized by dynamic interdependencies between physical assets. Traditional models fail to capture these intricate relationships as they treat buildings as independent entities or rely on static correlation structures. This paper introduces a novel Spatiotemporal Graph Neural Network (ST-GNN) framework that reframes infrastructure forecasting as a relational reasoning task, enabling dynamic inference of campus wide interdependencies. Our approach integrates Graph Attention Networks (GAT) to learn time-varying spatial dependencies and Gated Temporal Convolutional Networks (TCNs) to capture multi-scale temporal patterns. A key innovation is our context-sensitive graph construction method that incorporates physical proximity, functional similarity, and human mobility data to create a holistic representation of campus dynamics. Evaluated on a real-world multimodal dataset comprising 24 months of energy and occupancy data from 50 campus buildings, the proposed model demonstrates superior performance, achieving a 16.3% reduction in mean absolute error compared to the strongest baseline. Comprehensive ablation studies confirm the critical contribution of each architectural component, while qualitative analysis reveals the model’s capacity to provide interpretable insights into campus operational patterns. This work provides a powerful framework for intelligent campus management, enabling precise resource allocation, energy optimization, and sustainable operational planning through advanced relational reasoning capabilities.
Deep Learning Model for Interpretability and Explainability of Aspect-Level Sentiment Analysis Based on Social Media Nikhil Kumar Singh, Sanjay Agal, Thippa Reddy Gadekallu, Mohammad Shabaz, Ismail Keshta, Latika Jindal, Mukesh Soni, Haewon Byeon, Pavitar Parkash Singh IEEE Transactions on Computational Social Systems, 2025 The interactive attention graph convolution network (IAGCN), a novel model proposed in this article, will revolutionize aspect-level sentiment analysis (SA). IAGCN effectively addresses these key features, in contrast to prior research that ignored the meaning of aspect terms and their relationship with context. The model combines a modified dynamic weighting layer with bidirectional long short-term memory (BiLSTM) to accurately acquire context. It takes use of graph convolutional networks (GCNs) to encrypt syntactic information from the syntactic dependency tree. Furthermore, a method for interactive attention is employed to discover the intricate relationships between context and aspect terms, which results in the reconstruction of those terms’ representations. Comparing the proposed IAGCN model to baseline models, impressive gains are made. Across five datasets, the model beats previous methods with an amazing improvement in F1 scores that ranges from 1.34% to 4.04% and an impressive improvement in accuracy that ranges from 0.56% to 1.75%. Additionally, the IAGCN model outperforms the global vectors (GloVe)-based strategy when the potent pretrained model bidirectional encoder representations from transformers (BERT) is included in the challenge, resulting in even greater improvements. The F1 score considerably increases from 2.59% to 7.55%, and accuracy increases from 1.47% to 3.95%, making the IAGCN model a standout performer in aspect-level SA.
IoT as a Tool for Remote Engineering Education Opportunities and Challenges Sanjay Agal, Niyati Dhirubhai Odedra Iet Conference Proceedings, 2025 This paper investigates the role of Internet of Things (IoT) technologies in enhancing remote engineering education, with a focus on improving student engagement, learning outcomes, and skill development while addressing challenges such as accessibility, technological limitations, and pedagogical gaps. Employing a mixed-methods approach, the research incorporates qualitative and quantitative data gathered through surveys, interviews, and case studies from various educational institutions that have successfully integrated IoT into their remote learning frameworks. The findings reveal that IoT can significantly enhance interactivity and participation, leading to improved comprehension and retention of engineering concepts, while also highlighting persistent challenges including disparities in technological access and the necessity for effective pedagogical strategies tailored to IoT-enabled environments. This research is significant as it underscores the potential of IoT technologies to transform educational practices, particularly during times of increased reliance on remote learning modalities, such as those seen in healthcare education. By providing actionable insights into the effective implementation of IoT in educational settings, this study not only contributes to the advancement of engineering pedagogy but also offers broader implications for the healthcare field, suggesting that the integration of such technologies could facilitate the upskilling of future healthcare professionals and enhance collaborative learning experiences. Ultimately, this research advocates for enhanced policies and support structures that foster the effective use of IoT in education, ensuring equitable access and bridging existing gaps in learning outcomes.
A Comprehensive Study on Lattice, Code, and Hash-Based Cryptographic Algorithms in Post-Quantum Security with Practical Applications Mohammad Asif, Sanjay Agal Iet Conference Proceedings, 2025 A computer is recognized as a universal device that can simulate any other computational device, though it may take time. Cryptography is becoming more important as the Internet becomes more popular. When shopping online or using Internet banking, safe communication is crucial. However, traditional cryptography often relies on certain computational challenges, such as the RSA system, which is because of the complexity of factoring integers. This makes it vulnerable to advances in hardware, algorithms, and quantum techniques such as the Shor algorithm. In today's digital world, secure communication is essential. Quantum physics can change traditional cryptography. For example, an eavesdropper like Eve could record 2014 messages and, with future technology, decrypt them. Distribution of quantum keys (QKD) is a key application of quantum cryptography and promises unconditional communication security, unlike traditional methods that depend on unproven computational assumptions. Instead, QKD focuses on the physical laws. A challenge is the secure distribution of this key. In traditional methods, trusted couriers may be compromised unknowingly. QKD, introduced by Bennett and Brassard in 1984, involves Alice sending Bob photons with different polarization. They share their measurement results through a secure channel and generate a key while checking for interference. The onetime pad protocol ensures complete security if the key is not reused. QKD has become advanced enough for practical use. The article discusses cutting edge encryption methods including Lattice, Hash and Code-Based Cryptography, and their uses.
Soybean Leaf Pathogens Classification Using Fine-Tune Vision Transform Pranav Deepak Kavathekar, Kush Bhushanwar, Sanjay Agal Iet Conference Proceedings, 2025 The research creates a classification system for soybean leaf pathogens through a Vision Transformer (ViT) model with fine-tuning capabilities. Traditional convolutional neural networks (CNNs) remain popular for plant disease detection however Vision Transformers (ViTs) demonstrate better image feature extraction because they can process distant relationships between image features. Scientists conducted research using a ViT model which received pre-training afterward they applied it to diagnose soybean leaf images with different pathogen-related symptoms. The training process included data augmentation and application of the Adam optimizer as optimization methods to boosts generalizability. Experimental tests showed that the optimized ViT reached 99% accuracy demonstrating better performance than standard CNN-models. The conducted research demonstrates how ViT models present promising capabilities for precise and automated soybean disease detection that enables timely disease administration to enhance farming productivity.
A Review on Soybean Leaf Pathogens Classification Methodologies Pranav Deepak Kavathekar, Kush Bhushanwar, Sanjay Agal Proceedings of the International Conference on Multi Agent Systems for Collaborative Intelligence Icmsci 2025, 2025 Soybean leaf diseases are one of the major challenges to agriculture in many countries. However, it is advisable to detect and classify these diseases properly and on time. This review focuses on the recent studies concerning various approaches for soybean leaf diseases detection such as Machine Learning, Deep Learning, Transfer Learning. The goal of this paper is to propose a concise and algorithmic overview of the methods, the involved datasets, the means of preparing the data, and the metrics for evaluating the performance of these methods. This review discusses the achievements of the past, challenges of the present, and prospects of future detection of soybean leaf diseases. Its purpose is to assist researchers and practitioners to disseminate existing information and identify directions for improving disease detection systems to produce better innovative results more quickly.
A privacy preserving synthetic learner dataset for learning analytics in technology enhanced higher education S Agal Scientific Reports , 2026 2026 Citations: 1
A Machine Learning Architecture Integrating Spatiotemporal Graph Networks with Differentiable Optimization for Adaptive System Management S Agal, N Odedra PUXplore Multidisciplinary Journal of Engineering 2 (1), 1-58 , 2026 2026
A machine learning approach to risk based asset allocation in portfolio optimization S Agal, K Raulji, ND Odedra Scientific Reports 15 (1), 42263 , 2025 2025 Citations: 8
Optimizing Impact Investment Portfolios with Reinforcement Learning: A Data-Driven Framework for Balancing Financial Returns and SDG Alignment S Agal, K Raulji, K Shekokar, N Bhavsar FinTech and Sustainable Innovation, 1-15 , 2025 2025 Citations: 1
ICT Analysis and Applications: Proceedings of ICT4SD 2025, Volume 12 S Fong, N Dey, A Joshi Springer , 2025 2025 Citations: 4
Morlet Empowering Wavelet Decision Neural Network with Lotus Effect Optimization Algorithm based Classification of Skin Cancer K Thenmozhi, B Kulkarni, SP Velmurugan, S Agal, R Maranan, ... 2025 9th International Conference on Electronics, Communication and … , 2025 2025
Decentralized reinforcement learning for scalable embodied intelligence in robotic swarms S Agal, ND Odedra Emb Intell Robot , 2025 2025 Citations: 1
Spatiotemporal graph networks for relational reasoning in campus infrastructure management S Agal, K Raulji, N Bhavsar, P Bhatt Int. J. Adv. Comput. Sci. Appl 16 , 2025 2025 Citations: 4
A Data Analytics Framework for Measuring the Efficacy of Project Based Learning on SDG Focused FinTech Projects in Indian Universities S Agal, N Odedra PUXplore Multidisciplinary Journal of Engineering 1 (2) , 2025 2025
Data science in embodied artificial intelligence and robotics: A comprehensive study of models, methods, and applications S Agal, ND Odedra Embodied Intelligence and Robotics, 025200005 , 2025 2025
An Improved Link Prediction Algorithm for Financial Transaction Graphs Based on Structural Similarity and Firm Attributes S Agal, N Bhavsar, S Das 2025 World Skills Conference on Universal Data Analytics and Sciences … , 2025 2025
Integrating Edge Computing with Swarm Intelligence for Efficient IoT Device Management NB Muni, HK Bhargav, M Sarkar, S Nandi, S Agal, A Vasmatkar 2025 3rd International Conference on Data Science and Information System … , 2025 2025 Citations: 1
Improving Aerial Image Registration by Outlier Filtering Through Feature Classification AM Raju, K Kirubasankar, S Agal, S Gawshinde 2025 3rd International Conference on Data Science and Information System … , 2025 2025 Citations: 1
Enhancing Outcome Based Education with Data Science: A Case Study at Parul University S Agal, KM Raulji, P Parashar, K Shekokar International Conference on Digital Age & Technological Advances for … , 2025 2025
Heartbeat sensor fault detection in IoT using machine learning DM Bhatt, A Gandhi, S Agal IET Conference Proceedings CP920 2025 (7), 739-745 , 2025 2025
A comprehensive study on lattice, code, and hash-based cryptographic algorithms in post-quantum security with practical applications M Asif, S Agal Parul University International Conference on Engineering and Technology 2025 … , 2025 2025 Citations: 4
IoT as a tool for remote engineering education opportunities and challenges S Agal, ND Odedra Parul University International Conference on Engineering and Technology 2025 … , 2025 2025 Citations: 4
Revolutionizing eHealth and mHealth: integrating IoT, big data, and AI for enhanced patient monitoring MI Shaikh, D Chakraborty, SMM Naimoddin, S Agal Parul University International Conference on Engineering and Technology 2025 … , 2025 2025
Soybean leaf pathogens classification using fine-tune vision transform PD Kavathekar, K Bhushanwar, S Agal Parul University International Conference on Engineering and Technology 2025 … , 2025 2025
Improving Robustness and Accuracy in Unsupervised Domain Adaptation S Agal, D Pandey, N Bhavsar, NK Jain, P Bhatt International Conference on Information and Communication Technology for … , 2025 2025
MOST CITED SCHOLAR PUBLICATIONS
Deep learning model for interpretability and explainability of aspect-level sentiment analysis based on social media NK Singh, S Agal, TR Gadekallu, M Shabaz, I Keshta, L Jindal, M Soni, ... IEEE Transactions on Computational Social Systems 12 (3), 1307-1318 , 2024 2024.0 Citations: 32
Elevating Offensive Language Detection: CNN-GRU and BERT for Enhanced Hate Speech Identification. M Madhavi, S Agal, ND Odedra, H Chowdhary, TS Ruprah, VA Vuyyuru, ... International Journal of Advanced Computer Science & Applications 15 (5) , 2024 2024.0 Citations: 29
A study and overview on current trends and technology in mobile applications and its development H Rathod, S Agal International conference on ICT for sustainable development, 383-395 , 2023 2023.0 Citations: 24
NLP-Based Automatic Summarization using Bidirectional Encoder Representations from Transformers-Long Short Term Memory Hybrid Model: Enhancing Text Compression. RS Kartha, S Agal, ND Odedra, CSK Nanda, VS Rao, AM Kuthe, ... International Journal of Advanced Computer Science & Applications 15 (5) , 2024 2024.0 Citations: 22
Revolutionizing Data Capitalization: Harnessing Blockchain for IoT-Enabled Smart Contracts C Thingom, MR Tammina, A Joshi, S Agal, MSI Sudman, H Byeon 2023 Second International Conference On Smart Technologies For Smart Nation … , 2023 2023.0 Citations: 20
A Novel Study on Data Science for Data Security and Data Integrity with Enhanced Heuristic Scheduling in Cloud NSA Polireddi, M Suryadevara, S Venkata, S Rangineni, SKR Koduru, ... 2023 2nd International Conference on Automation, Computing and Renewable … , 2023 2023.0 Citations: 18
Broadcast speech recognition and control system based on Internet of Things sensors for smart cities M Qin, R Kumar, M Shabaz, S Agal, PP Singh, A Ammini Journal of Intelligent Systems 32 (1), 20230067 , 2023 2023.0 Citations: 17
The Analytical CRM OLAP Analysis Tools and Data Mining S Agal, P Devija ICT Analysis and Applications: Proceedings of ICT4SD 2019, Volume 2, 1-7 , 2020 2020.0 Citations: 16
Available Bandwidth Estimation in MANET Using FPECM-MFL-GRRSU for Adaptive Video Streaming S Agal International Conference on ICT for Sustainable Development, 179-192 , 2023 2023.0 Citations: 15
MV, & Arri, HS (2023) S Agal, P Sharma, CR Mohan, P Madan Using Machine Learning Algorithms to Suggest a Method for Predictive … , 2023 2023.0 Citations: 15
IoT-Enabled Cloud-Based Fair Provable Data Possession Scheme based on Blockchain H Byeon, H Kaur, S Agal, S Kumar, M Manu, R Maranan 2023 Second International Conference On Smart Technologies For Smart Nation … , 2023 2023.0 Citations: 13
Bandwidth estimation and optimized bitrate selection for dynamic adaptive streaming over HTTP using RSI-GM and ISSO S Agal, PK Gokani International Journal of Computer Vision and Image Processing (IJCVIP) 12 (1 … , 2022 2022.0 Citations: 9
A machine learning approach to risk based asset allocation in portfolio optimization S Agal, K Raulji, ND Odedra Scientific Reports 15 (1), 42263 , 2025 2025.0 Citations: 8
An optimized bandwidth estimation for adaptive video streaming systems using WLBWO algorithm S Agal, PK Gokani International Journal of Interdisciplinary Telecommunications and Networking … , 2021 2021.0 Citations: 8
Factors Influencing the patients in attaining satisfaction by the services provided in the hospital P Devija, S Bhandari, S Agal International Journal of Management & Business Studies 2 (3), 95-98 , 2012 2012.0 Citations: 8
A unified framework for smart and secure digital transformation: Leveraging block chain, ai, and ict across healthcare, education, e-commerce, and industrial systems S Agal, N Bhavsar, K Shekokar, SS Shrivastava Int. J. Sci. Res. Arch 15 (1), 927-943 , 2025 2025.0 Citations: 7
SVM Modeling Simulation to Evaluate the Electric Vehicle Transmitting Points RK Kaushal, S Agal, R Singh, PP Singh 2023 International Conference on Advances in Computing, Communication and … , 2023 2023.0 Citations: 6
PRINCIPLES AND PRACTICES OF NETWORK SECURITY IR Khan, M Sandhu, S Agal, HN Patel Xoffencerpublication , 2023 2023.0 Citations: 6
Dr. L. Sridhara Rao, Dr. Sanjay Agal, & Dr. Haewon Byeon.(2023). THE ART OF INTELLIGENT MACHINES UNLEASHING THE POWER OF MACHINE LEARNING MOP Singh THE ART OF INTELLIGENT MACHINES UNLEASHING THE POWER OF MACHINE LEARNING, 214 , 0 Citations: 6
Multidisciplinary AI and data science applications in fintech: A case study from Parul University S Agal, N Bhavsar, K Raulji, K Shekokar International Journal of Latest Technology in Engineering Management & … , 2025 2025.0 Citations: 5
