Design of A Recurrent Neural Network Model for Machine Reading Comprehension

Reading paragraphs, understanding the related questions and answering them has always been a difficult task for the machines. Humans have the capability to understand the logic and meaning of a question and answer it to maintain a proper mode of interaction. But for machines, this is a complex task. The main focus of this research work is to explore the various machine learning and neural networks based techniques to develop and train a model on context of paragraphs related questions and answers and then test the model on an user given paragraph and question. This paper presents a thorough understanding of data analysis of the SQuAD data set and word embedding applied on the questions and answers of training set of the SQuAD data set. (C) 2020 The Authors. Published by Elsevier B.V.