An Overview of Technological Revolution in Deep Learning Architectures for Biomedical Named Entity Recognition

The research publications related to biomedical domain are increasing and a huge information is concealed in it. Most of these information are available in the unstructured format and it is not available in ready-to-use format for further experiments and analysis. Several techniques and features of Information Extraction (IE) and Natural Language Processing (NLP) in Text Mining (TM) are used in the field of biomedical domain. The main purpose of this paper is to review the various research works / practices which are being carried out, especially in the Biomedical Named Entity Recognition (BioNER) domain. Application of NLP in BioNER helps to extract knowledge on various entities like drug, disease, chemical, cell, gene, protein, etc. In this paper, a survey on BioNER which includes traditional rule-based methods, statistical Machine Learning (ML) and Deep Learning (DL) has been carried out to get key insights on research works in this area. This review proves that the DL models took over statistical ML and rule-based systems and proved to be the state-of-the-art in various perspectives.