A Personalized and Privacy-Preserving Federated Transformer Framework for Multilingual Sentiment Analysis Jothi Prakash V, Arul Antran Vijay S, Gopikrishnan Sundaram IEEE Transactions on Pattern Analysis and Machine Intelligence, 2026 Personalized federated learning for multilingual sentiment analysis poses significant challenges arising from linguistic heterogeneity, non-IID data distributions, and strict privacy requirements. This paper proposes FedPerX, a federated transformer framework that integrates residual adapter-based personalization with adaptive multi-granular differential privacy. The architecture leverages a frozen multilingual backbone (XLM-R) while enabling each client to train lightweight, client-specific adapters. Privacy is enforced through dynamic noise injection at both the feature and adapter levels, calibrated using gradient sensitivity. FedPerX is evaluated on two multilingual benchmarks-MARC and TSMD-spanning structured reviews and informal social media content across more than ten languages. Experimental results demonstrate consistent improvements over seven state-of-the-art baselines, with up to +4.3% gains in macro-F1, a 70% reduction in communication overhead, and the lowest variance in client-level performance. Comprehensive analyses, including fairness, personalization gap, privacy-utility trade-off, and ablation studies, validate the framework's robustness and adaptability. FedPerX advances the design of scalable, personalized, and privacy-preserving models for federated multilingual sentiment analysis.
Toward Explainable Cross-Lingual Adaptive NAS for Enhanced Tamil Medical Text Summarization Jothi Prakash V, Arul Antran Vijay S, Gopikrishnan Sundaram IEEE Journal of Biomedical and Health Informatics, 2026 In the era of digital health, the demand for effective summarization of medical texts in low-resource languages, such as Tamil, is rapidly increasing. Traditional neural models often struggle with this task due to the scarcity of annotated data and the complexity of medical terminology. To address these challenges, we propose a novel Cross-Lingual Adaptive Neural Architecture Search (CLANAS) framework, specifically designed to enhance Tamil medical text summarization by leveraging cross-lingual transfer learning. The CLANAS framework integrates embedding alignment techniques with neural architecture search (NAS) to automatically design optimal models tailored to the target language. Our methodology involves pre-training on large-scale English medical datasets, including iCliniq, HealthCare Magic, MeQSum, MEDIQA, and BioLaySumm-2023, followed by fine-tuning on Tamil medical texts, ensuring semantic consistency across languages through embedding alignment. The framework was rigorously evaluated on these benchmark datasets, where CLANAS demonstrated significant performance improvements, achieving up to 9.3% enhancement in ROUGE-1 scores, 8.4% in BLEU, and 7.5% in Metric for Evaluation of Translation with Explicit ORdering (METEOR) compared to state-of-the-art models. Ablation studies further confirmed the effectiveness of each component within the framework. These results underscore the potential of CLANAS as a robust solution for improving the quality of medical text summarization in Tamil.
The Convergence of Deep Learning and the Metaverse: A Multidisciplinary Survey of Current Research and Future Directions Jothi Prakash Venugopal, Arul Antran Vijay Subramanian, Gopikrishnan Sundaram, Mahendhiran Ponnambalam Devadoss Computer Animation and Virtual Worlds, 2025 The convergence of deep learning and the Metaverse represents a pivotal frontier in the evolution of intelligent digital ecosystems. This paper presents a comprehensive survey of how deep learning techniques spanning convolutional, generative, transformer‐based, and reinforcement architectures collectively enable perception, creation, cognition, and governance within immersive virtual worlds. Building upon this synthesis, we propose the Deep Learning‐Empowered Metaverse Intelligence (DL‐MI) framework, which unifies sensory intelligence, generative world‐building, adaptive reasoning, and ethical‐social governance into a cohesive architecture. The study illustrates how deep learning facilitates realistic avatar synthesis, dynamic environmental rendering, emotion‐aware interaction, and predictive personalization, thereby transforming the Metaverse from reactive systems to anticipatory, self‐evolving spaces. Key challenges such as data privacy, algorithmic bias, and computational sustainability are critically examined alongside emerging paradigms, including quantum‐augmented AI and federated collaboration. By integrating technical, ethical, and societal dimensions, this survey provides a structured foundation for developing scalable, transparent, and human‐centered Metaverse intelligence.
A Graph-Based Framework for Temporal and Causal Analysis of Sentiments Subha E, Jothi Prakash V, Arul Antran Vijay S ACM Transactions on the Web, 2025 This research aims to develop a novel framework that uncovers the causal influence of global events on public sentiment through temporal graph modeling and neural causal inference. Global events, such as pandemics, elections, and economic crises, profoundly affect public sentiments, shaping social behaviors and economic outcomes. Traditional models often fall short in capturing the complex, dynamic, and non-linear relationships between these events and sentiments. This article presents the Neural Temporal Causal Graph Network (NTCGN), a unified framework that integrates temporal graph neural networks with a Causal Attention Network (CAN) to model and interpret these relationships. NTCGN constructs a temporal graph from event data and sentiment-labeled texts, learning dependencies and causal influences through advanced neural architectures. A thorough comparative analysis with state-of-the-art models such as Logistic Regression, SVM, LSTM, and transformer-based models demonstrates NTCGN’s superior performance. Experimental evaluation using the Sentiment140 and Global Database of Events, Language and Tone (GDELT) 2.0 datasets shows NTCGN achieving an accuracy of 0.798 and an F1 score of 0.795, outperforming these baseline models. The model’s causal inference capabilities are validated using the Causal Impact Score (CIS) and Causal Discovery Precision (CDP), highlighting its reliability in identifying true causal links. Visualizations of attention maps and causal pathways enhance interpretability, demonstrating how specific events influence public sentiments. This work provides a robust and interpretable tool for analyzing event-driven sentiment dynamics in real-world applications.
A Comprehensive Multimodal Framework for Optimizing Social Media Hashtag Recommendations Jothi Prakash V, Arul Antran Vijay S IEEE Transactions on Computational Social Systems, 2025 In the dynamic landscape of social media, the strategic use of hashtags has emerged as a crucial tool for enhancing content discoverability and engagement. This research introduces the neurosymbolic contrastive framework (NSCF), an innovative methodology designed to address the multifaceted challenges inherent in automated hashtag recommendation, such as the integration of multimodal data, the context sensitivity of content, and the dynamic nature of social media trends. By combining deep learning's representational strengths with the deductive prowess of symbolic artificial Intelligence (AI), NSCF crafts contextually relevant and logically coherent hashtag suggestions. Its dual-stream architecture meticulously processes and aligns textual and visual content through contrastive learning, ensuring a comprehensive understanding of multimodal social media data. The framework's neurosymbolic integration leverages structured knowledge and logical inference, significantly enhancing the relevance and coherence of its recommendations. Evaluated against a variety of datasets, including MM-INS, NUS-WIDE, and HARRISON, NSCF has demonstrated exceptional performance, outshining existing models and baseline methods across key metrics such as precision (0.721–0.701), recall (0.736–0.716), and F1 score (0.728–0.708). This research represents a major advancement in social media analytics as it not only demonstrates NSCF's novel approach but also sheds light on its potential to transform hashtag recommendation systems.
EFFECTS OF CUTTING AND WELDING ON ELECTRIC STEELS' FERROMAGNETIC MATERIAL Abishek Agarhari, Thakur Singh, Meenu, N. V. Balaji, S. Arul Antran Vijay, Monali Ravindra Borade Proceedings of International Conference on Sustainable Computing and Integrated Communication in Changing Landscape of AI Icscai 2024, 2024
Segmentation Based on 1-Shot Learning A. Senthil Kumar, Ritika Mehra, Prashant Kumar, S. Hemelatha, S. Arul Antran Vijay, P.R. Rege Proceedings of International Conference on Contemporary Computing and Informatics Ic3i 2023, 2023
Deep Neural Network Deception Using A Single Pixel Assault Digvijay Singh, Annu Yadav, Rishabh Arora, G. Ravivarman, S. Arul Antran Vijay, Y. K. Sharma Proceedings of International Conference on Contemporary Computing and Informatics Ic3i 2023, 2023
A hybrid colony fuzzy system for analyzing diabetes microarray data P. Ganesh Kumar, S. Arul Antran Vijay, D. Devaraj Proceedings of the IEEE Symposium on Computational Intelligence in Bioinformatics and Computational Biology Cibcb 2013 2013 IEEE Symposium Series on Computational Intelligence Ssci 2013, 2013
RECENT SCHOLAR PUBLICATIONS
A transformer-based uncertainty quantification framework for multimodal financial sentiment analysis E Subha, VJ Prakash, SAA Vijay International Journal of Machine Learning and Cybernetics 17 (3), 128 , 2026 2026
A novel approach to cross-linguistic transfer learning for hope speech detection in Tamil and Malayalam VJ Prakash, S Vijay, A Antran COMPUTER SPEECH AND LANGUAGE 96 , 2026 2026
The Convergence of Deep Learning and the Metaverse: A Multidisciplinary Survey of Current Research and Future Directions JP Venugopal, AAV Subramanian, G Sundaram, ... Computer Animation and Virtual Worlds 36 (6), e70082 , 2025 2025 Citations: 3
A Graph-Based Framework for Temporal and Causal Analysis of Sentiments E Subha, VJ Prakash, S Vijay, A Antran ACM TRANSACTIONS ON THE WEB 19 (4) , 2025 2025 Citations: 1
An integrated framework for emotion and sentiment analysis in Tamil and Malayalam visual content: Jothi Prakash V, Arul Antran Vijay S VJ Prakash, SAA Vijay Language Resources and Evaluation 59 (3), 2103-2141 , 2025 2025 Citations: 5
Advanced Seizure Detection with Convolutional Gated Recurrent Neural Networks: A Comprehensive EEG Signal Investigation S Fiza, SAA Vijay, B Juneja, AK Keshari, R Jain 2025 IEEE Madhya Pradesh Section Conference (MPCON), 595-600 , 2025 2025
Sector-Based Encoding Schemes for Transition Reduction and Power Optimization in VLSI Systems NM Nandhitha, SAA Vijay, A Dev, S Kumar, A Kaur, N Kumar 2025 IEEE Madhya Pradesh Section Conference (MPCON), 451-456 , 2025 2025
Enhanced Privacy-Preserving and Trust Model for Secure Access Control and Data Transmission in Edge Computing A Rajiv, S Safinaz, SAA Vijay, P Madan, G Sahu, M Gupta 2025 World Skills Conference on Universal Data Analytics and Sciences … , 2025 2025
Optimized ECG Segmentation Using Hilbert Transform for Accurate Signal Classification A Gupta, A Panda, A Devan, SAA Vijay 2025 World Skills Conference on Universal Data Analytics and Sciences … , 2025 2025
A Novel Approach to ECG Signal Classification Using Enhanced Crow Search Algorithm for Feature Extraction SAA Vijay, A Sachdeva, PK Jena, S CV 2025 World Skills Conference on Universal Data Analytics and Sciences … , 2025 2025 Citations: 1
Multi-Tier Linguistic and Emotional Modeling for Cyberbullying Detection in Tamil Social Media VJ Prakash, SAA Vijay Expert Systems with Applications, 129270 , 2025 2025 Citations: 5
A Framework for Secure, Energy-Efficient Smart Buildings: Dynamic Security and Efficiency Enhancements SAA Vijay, D Sharma, SP Tripathy, B Bhushan, DB Rathod 2025 2nd International Conference On Multidisciplinary Research and … , 2025 2025
DCAT: A novel transformer-based approach for dynamic context-aware image captioning in the tamil language JP Venugopal, AAV Subramanian, M Murugan, G Sundaram, M Rivera, ... Applied Sciences 15 (9), 4909 , 2025 2025 Citations: 3
Elderly Healthcare Using Federated Learning Approach S Arul Antran Vijay, J Jebson, A Subairkhan, A Vasanthakumar, ... International Conference on Computing and Machine Learning, 665-677 , 2025 2025
Fuzzy Logic Classifier Approach for Non-Proliferative Diabetic Retinopathy in Macular Disease Quantification and Diagnosis T Bernatin, SAA Vijay, P Sharma, MH Mahalat, G Sood, X VK 2025 7th International Conference on Signal Processing, Computing and … , 2025 2025
Explainable AI For Leukemia Diagnosis: Interpreting Optimized Deep Learning Models for Improved Clinical Decision-Making R Reena, V Ojha, B Debnath, SK Samarwar, SAA Vijay, N Sharma 2025 IEEE International Conference on Emerging Technologies and Applications … , 2025 2025
A novel arctic fox survival strategy inspired optimization algorithm E Subha, V Jothi Prakash, SA Antran Vijay Journal of Combinatorial Optimization 49 (1), 1 , 2025 2025 Citations: 4
The New Era of Cartography with Visual Analytics and Augmented Reality Transforming the Field of Digital Globes L Dhingra, KL Maney, S Habbu, SAA Vijay, S Gupta, S Yuvaraj 2024 IEEE 4th International Conference on ICT in Business Industry … , 2024 2024
Advanced Techniques for Detecting Intrusions Leveraging Ensemble Machine Learning and Feature Reduction Techniques J Deepak, VA Mishra, SAA Vijay, A Mishra, K Karthika, M Nagpal 2024 IEEE 4th International Conference on ICT in Business Industry … , 2024 2024
Silicon Celebrations Merging with Systematic Structures to Shape Tomorrow’s Tech Landscape RG Anand, GB Deshmukh, SAA Vijay, A Mishra, V Mahalakshmi, P Garg 2024 IEEE 4th International Conference on ICT in Business Industry … , 2024 2024
MOST CITED SCHOLAR PUBLICATIONS
The realm of metaverse: A survey JP Venugopal, AAV Subramanian, J Peatchimuthu Computer Animation and Virtual Worlds 34 (5), e2150 , 2023 2023 Citations: 91
A novel attention-based cross-modal transfer learning framework for predicting cardiovascular disease NK Jothi Prakash V, Arul Antran Vijay S, Karthikeyan Computers in Biology and Medicine 170, 107977 , 2024 2024 Citations: 53
A context-sensitive multi-tier deep learning framework for multimodal sentiment analysis A Arul Antran Vijay S, Paul, A Nayyar Multimedia Tools and Applications 83 (18), 54249-54278 , 2024 2024 Citations: 44
Fuzzy expert system based on a novel hybrid stem cell (HSC) algorithm for classification of micro array data SAA Vijay, P GaneshKumar Journal of medical systems 42 (4), 61 , 2018 2018 Citations: 37
A deep ensemble network model for classifying and predicting breast cancer AAV Subramanian, JP Venugopal Computational Intelligence 39 (2), 258-282 , 2023 2023 Citations: 29
A multi-aspect framework for explainable sentiment analysis V Jothi Prakash, S Arul Antran Vijay Pattern Recognition Letters 178, 122-129 , 2024 2024 Citations: 27
Cross-lingual sentiment analysis of Tamil language using a multi-stage deep learning architecture J Prakash, AAS Vijay ACM Transactions on Asian and Low-Resource Language Information Processing … , 2023 2023 Citations: 21
An Automated Technology for IoT based Rail-Track Inspection to Locate Surface Flaws by Robotics and Neural Networks BU Maheswari, P Nithya, S Vijay, K Tamilarasi, N Muthukumaran 2022 International Conference on Inventive Computation Technologies (ICICT … , 2022 2022 Citations: 21
A comprehensive approach to bias mitigation for sentiment analysis of social media data JP Venugopal, AAV Subramanian, G Sundaram, M Rivera, P Wheeler Applied Sciences 14 (23), 11471 , 2024 2024 Citations: 20
A hybrid colony fuzzy system for analyzing diabetes microarray data PG Kumar, SAA Vijay, D Devaraj 2013 IEEE Symposium on Computational Intelligence in Bioinformatics and … , 2013 2013 Citations: 14
The impact of industrial automation on the manufacturing industry in the era of industry 4.0 VS Gurav, A Gugnani, YR Meena, V Marathe, SAA Vijay, S Nanda 2024 15th International Conference on Computing Communication and Networking … , 2024 2024 Citations: 12
A novel socio-pragmatic framework for sentiment analysis in Dravidian-English code-switched texts VJ Prakash, S Vijay, A Antran Knowledge-Based Systems 300 , 2024 2024 Citations: 9
A unified framework for analyzing textual context and intent in social media VJ Prakash, SAA Vijay ACM Transactions on Intelligent Systems and Technology 15 (6), 1-25 , 2024 2024 Citations: 7
A comprehensive learning based swarm optimization approach for feature selection in gene expression data S Easwaran, JP Venugopal, AAV Subramanian, G Sundaram, B Naseeba Heliyon 10 (17) , 2024 2024 Citations: 7
A modified firefly deep ensemble for microarray data classification AAS Vijay, J Prakash COMPUTER JOURNAL 65 (12), 3265-3274 , 2022 2022 Citations: 7
2022 International Conference on Inventive Computation Technologies (ICICT) BU Maheswari, P Nithya, S Vijay, K Tamilarasi, N Muthukumaran Nepal , 2022 2022 Citations: 6
Spectral analysis based differentiationfor EEG signals of children with autism B Arunkumar, SAA Vijay, KS Kumar J. Crit. Rev 7 (4), 882-887 , 2020 2020 Citations: 6
An integrated framework for emotion and sentiment analysis in Tamil and Malayalam visual content: Jothi Prakash V, Arul Antran Vijay S VJ Prakash, SAA Vijay Language Resources and Evaluation 59 (3), 2103-2141 , 2025 2025 Citations: 5
Multi-Tier Linguistic and Emotional Modeling for Cyberbullying Detection in Tamil Social Media VJ Prakash, SAA Vijay Expert Systems with Applications, 129270 , 2025 2025 Citations: 5
A novel arctic fox survival strategy inspired optimization algorithm E Subha, V Jothi Prakash, SA Antran Vijay Journal of Combinatorial Optimization 49 (1), 1 , 2025 2025 Citations: 4
