基于自动问答系统的信息检索技术研究进展

Survey on information retrieval system based on question answering system

自动问答是根据用户以自然语言提出的问题给出一个明确的答案。近年来,自动问答越来越受到信息检索和自然语言处理的研究者的关注。典型的自动问答系统通常包含问题分析、文段检索和答案选择等部件。介绍了自动问答的最新研究进展和相关国际会议情况,着重阐述问题分类、查询扩展、文段检索和答案选择这四个热点技术的主要功能和常用方法,最后提出存在的一些问题和展望。

Question Answering （QA） aims to find actual answers to users＇ questions in natural language, It has attracted more and more attention from the researchers in information retrieval and natural language research field. A typical QA system adopts a pipeline structure that contains ＂question analysis＂, ＂passage retrieval＂ and ＂answer selection＂ modules. In this paper, the research literature and the famous international conferences in QA research area were surveyed. The functions and methods of hot topics were mainly presented such as question analysis, query expansion, passage retrieval and answer selection. Furthermore, some existing problems were proposed in these studies.