Semantic Relations Extraction in Biomedical Text Using Deep Learning Techniques
Deep Learning is an emerging field of machine learning that has several use cases. Use of deep learning in the biomedical field is one such use case. Semantic relation- ship extraction is a natural language technique that seeks to identify the relationship between entities in a piece of text. The proposed research is aimed towards understanding the semantic relationship between diseases and treatment in a given biomedical sentence. The state-of-the-art model to achieve the goal has been done with the help of machine learning techniques, obtaining an accuracy of 90.72% for 3-class classification . In this paper, an initiative has been taken to do the same with the help of deep learning approaches. This paper aims at identifying semantic relationships in biomedical text and various deep learning architectures using Rosario and Hearst dataset. The train accuracy obtained for a 9-class classification is 82.21%, whereas for a balanced binary classification is 98.24%.