Machine Reading Comprehension Using Bi-directional LSTM Sequence Model

S. Anbukkarasi, G. Suganya, J. Sakunthala and Suguna Angamuthu

Machine Reading Comprehension (MRC) is creating a system that could read a document, understand the question and answer accordingly. It focuses on building frameworks that consequently answer questions posed by human in a natural language. A computer understanding of natural language comprises of the ability of a program framework to make an interpretation of sentences into an inward portrayal with the goal of creating substantial responses to questions asked by a user. This paper discusses the performance of Bi-directional LSTM (Long Short- Term Memory) sequence model where the Bidirectional LSTM keeps the contextual information in both directions which is pretty useful in text classification task. The machine is trained with pair of comprehension and question answers which in turn would predict the answers for the questions that have been posted later.

Volume 12 | 03-Special Issue

Pages: 1501-1506

DOI: 10.5373/JARDCS/V12SP3/20201403