摘要
在汉语问答系统中,当用自然语言问句进行文档检索时,由于问句比查询词包含更多的语义信息,因此必须进行查询词扩展以提高信息检索的性能.通过分析已有的查询扩展方法,提出了基于集合论的查询扩展新方法.它结合了3种传统的查询扩展方法:语义词典法、自动相关反馈法和问题类型词.实验结果表明该方法在Web检索方面是有效并且优于传统的方法.
In Chinese question answering system, while using natural language questions to retrieve documents, because of more abundant semantic relation of question than that of query words, the precision can be improved by expanding query. On the basis of previous approaches to query expansion, this paper proposes a new approach to query expansion based on set theory, which combines three traditional methods--thesauri, automatic relevance feedback and the related words of the question type. Experiments show the new approach is effective to query expansion for web retrieval and outperforms the optimized, conventional expansion approaches.
出处
《江西师范大学学报(自然科学版)》
CAS
北大核心
2008年第2期211-214,共4页
Journal of Jiangxi Normal University(Natural Science Edition)
关键词
中文问答系统
查询扩展
问题类型词
相关反馈
语义词典
集合论
Chinese question answering system
query expansion
related words of question type
relevance feedback
thesauri
set theory