Bridging the Gap Between Question and Answer: A Comprehensive Analysis on Medical Question Answering Systems

Abstract

Data is the universal language of information in real world. But according to a statistic only 20% of the data in real world are structured where remaining 80% of data are unstructured. The machine provides in accurate result in retrieving the information from these unstructured data. With help of Natural Language Processing (NLP), the machines are able to process the dark data and accomplish the task given by user. Among many tasks of NLP, Question Answering System (QAS)plays a vital role for the real-world development. QAS is the task of giving accurate answer for the question posted by user in natural language about the document (Textual Question Answering) or query about the image (Visual Question Answering) or question related to medical field (Medical Question Answering). This paper provides an overview of Medical QAS Datasets, Methodology implemented and the Metrics to evaluate the model. At the end of the survey, this paper provides a finalized overview of what methodology/approach can be used for the QAS.