Semantic Relations Extraction in Biomedical Text Using Deep Learning Techniques

  • Ayush Eshan,Srilekha Vinjamara,Srinivasan R

Abstract

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  [14]. 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%.

Published
2020-06-01
How to Cite
Ayush Eshan,Srilekha Vinjamara,Srinivasan R. (2020). Semantic Relations Extraction in Biomedical Text Using Deep Learning Techniques. International Journal of Advanced Science and Technology, 29(7s), 4661 - 4671. Retrieved from https://sersc.org/journals/index.php/IJAST/article/view/25709