受限领域中文问答系统问句分析研究  

Study on question parsing of restricted-domain Chinese question answering system

作　　者：蒋昌金[1] 彭宏[1] 马千里[1] 林正春[1] 王成[1] 

机构地区：[1]华南理工大学计算机科学与工程学院,广东广州510006

出　　处：《计算机工程与设计》2010年第11期2588-2591,共4页Computer Engineering and Design

基　　金：广东省自然科学基金项目(07006474);广东省科技攻关基金项目(2007B010200044)

摘　　要：问句分析是问答系统中一个极其重要的部分,它对后续的搜索模块和答案抽取模块有着至关重要的影响。为了提高问句分析模块的性能,对中文问答系统问句分析模块进行研究和分析,在构建领域词典的基础上实现了对问句中命名实体的识别。对问句类型的识别则采用问题类别对照表的方式来实现。在对问句进行分词、词性标注之后,提取出其中的名词、动词、形容词等实词,形成初始查询向量,最后对初始查询向量进行同义词扩展和问句类型关联词扩展。实验结果表明,通过查询向量进行查询比用问句直接查询有着更高的准确率。Question parsing is a very important part of a question answering system and it has significant effect on the following searching module and answer extracting module.First profound research and analysis has been made for the question parsing module of a Chinese question answering system,and then the construction of the domain dictionary which used to Chinese named entity recognition is described.In order to identify the type of a question,a question type list is constructed.After parsing and segmenting a question,the nouns,verbs and adjectives in it are extracted.These words extracted from the question formed an initial query vector.This initial query vector is extended to an extended query vector by adding synonyms of the nouns and verbs in it and associate words of its type.The experimental results show that the query using query vector has a higher precision than the query using original question.

关 键 词：中文问答系统 问题分析 查询向量 查询扩展 信息检索 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]