Biography: Dr. T. Velmurugan is working as an Associate Professor in the PG and Research Department of Computer Science and Applications, Dwaraka Doss Goverdhan Doss Vaishnav College, Chennai-600106, India. Also, he is the Advisor and Head, Department of Computer Applications (BCA). He holds a Ph.D. degree in Computer Science from the University of Madras. He has 27 years of teaching experience. He guided more than 300 M.Phil., Research Scholars. Also, he guided 16 Ph.D. scholars and currently guiding 7 Ph.D. scholars. He has published more than 110 articles in SCOPUS and SCI indexed journals. He elected and served as a Senate Member from Academic Council, University of Madras. Also, he served as a nominated Senate Member in the Middle East University, Dubai, UAE for a period of three years. He has a lot of administrative experiences. He served as advisory board member to many academic institutions in and around Tamil Nadu, India. He was an invited speaker and keynote speaker for many international conferences around the world. He is a member in Board of studies for many autonomous institutions and Universities. Also, he organized international Conferences and workshops. In addition, he was a resource person for various national workshops entitled "Scientific Research Article Writing and Journal Publications" and many of the recent topics in Computer Science. He is an Editorial Board Member of 6 International Journals. He also a reviewer in many peer reviewed journals like Elsevier, Springer, IOSPress Journals etc. He is the Chair person for the Government of Tamil Nadu State for XII standard book titled as “Electronics and Hardwares”. Further, he is a visiting faculty for M.Phil. course for various universities throughout India. His H index is 18 and i10 index is 28. His area of specialization includes Data Mining, Artificial Intelligence, Machine Learning, Network Security, Big Data Analytics, Data Science and etc.
Speech Title: Semantic Relation Extraction in Biomedical Text Data Using Transfer Learning Techniques
Abstract: Relation Extraction (RE) is one of the most important research areas in the field of biomedical text. Early, most of the biomedical RE is based on Machine Learning (ML) approach. Recent progress of biomedical RE models was made possible by the advancements of deep learning (DL) techniques used in natural language processing. Although DL is better than ML in overall comparisons, it often requires an annotated corpus to train a new model. As well, most of them are mainly evaluated on annotated database and have not yet been broadly executed on unannotated databases. From a deep learning perspective, the RE classification problem can be solved through transfer learning. Transfer learning is a subclass of traditional machine learning approach, which transfers their knowledge from previously learnt domains to newer domains and tasks. The aim of this research work is to implement recent transfer learning algorithm using various biomedical text databases. In this sense, this is to introduce a transfer learning approach by BiLSTM (Bidirectional long short-term memory) and BioBERT (Bidirectional Encoder Representations from Transformers for Biomedical Text Mining) a pre trained language model to address the lack of high-quality, large-scale annotated biomedical data. It is evaluated on three important biomedical relation tasks: drug-disease association, drug-drug interaction, protein-protein interaction. In this research work, to find the accuracy of model is calculated using standard measures. The experimental results show that the BiLSTM-BioBERT achieved state-of-the-art performance. Therefore, the proposed work is suggested for better use in biomedical data. Finally this work can be used to understand complex biomedical texts and provides strong baselines for future research.