Font Size: a A A

Design And Implementation Of XJAUQA Automatic Question Answering System

Posted on:2007-05-19Degree:MasterType:Thesis
Country:ChinaCandidate:X LiFull Text:PDF
GTID:2178360185452003Subject:Agricultural mechanization project
Abstract/Summary:PDF Full Text Request
Automatic Question Answering technology is developing with meeting people's desire of obtaining useful information fast and accurately. The system of automatic Question Answering (Q/A) can return the concise answers to the problem of users in natural language style. Currently Automatic Question Answering system based Frequently Asked Questions system (FAQ) is one of the three big research fields which study Automatic Question Answering system, and the most important application for its practical significance, too. It is this paper's emphases.In the modern distant education and the network teaching, the automatic Question Answering system, which is more practical and easy to develop and be maintained, is always aim at some concrete curriculum, therefore the questions proposed by the users are also relatively concentrated. We organize these questions to form the FAQ stored in the system, thereby we can enhance the accuracy and perform efficiency.This paper describes the design and implementation of a Chinese Natural Language automatic Question Answering system XJAUQA, which makes answers to subjective questions of special curriculum used in network teaching environment. XJAUQA avoids a fundamental problem of synonym and polyseme in the conventional retrieval system by importing retrieval precision by Latent Semantic Index technology. The system adopts B/S pattern, which combine Latent Semantic Index technology and the website development technology; mainly depend on natural language automatic Question Answering, keyword retrieve and teacher on-line explanation auxiliary, fully display Internet education superiority and enhance teaching proficiency and decrease teacher's teaching burden.
Keywords/Search Tags:automatic Question Answering, FAQ, Latent Semantic Index, Information Retrieval
PDF Full Text Request
Related items