Design and Implementation of Automatic Question Answering System in Information Retrieval

Abstract

Q & A is an indispensable part in the teaching process. Developing a dialogue-based automated question answering system can answer students' questions accurately and assist the effective teaching of the course. Information retrieval has become one of the core courses in information technology education in major universities in China. However, there are still some problems in the current information retrieval course. This study takes the course of information retrieval as an example to design and implement a QA system based on Lucene. This system mainly includes four core function modules of data set construction, keyword extraction, problem retrieval and answer extraction. The workflow of the system is as follows: First, the system segments the textual content entered by the user into words and extracts the keywords. Then it would retrieve the keywords in the FAQ. Later, the system would extract the matching question and corresponding answer from the retrieved questions, and finally return the answer to the user. The development of QA system is based on the B/S architecture, which is finally presented to the user in the form of web pages. At the same time, the responsive design allows the system to render well in multiple terminals. Finally, the function test and performance test of the system are also carried out. The result shows that after the user enters the question, the system can return the relevant answer, and the accuracy rate is 0.84, which indicates that the automatic question answering system can basically meet the needs of users in both function and performance.