Rule based architecture for medical question answering system

Abstract

As the wealth of online information is increasing tremendously, the need for question-answering systems is evident. Current search engines return ranked lists of documents to the users query, but they do not deliver the precise answer to the queries. The goal of a question-answering system is to retrieve answers to questions rather than full documents or best-matching passages, as most information retrieval systems currently do. Patients/Medical students have many queries related to the medical terms, diseases, and its symptoms. They are inquisitive to find these answers using search engines. But due to keyword search used by search engines it becomes quite difficult for them to find the correct answers for the search item. This paper proposes the architecture of question-answering system for medical domain and discusses the rule-based question processing and answers retrieval. Rule formation for retrieval of Answers has also been discussed in the paper.