摘要
随着Internet的不断发展,现已成为全球最大的共享信息基地,与此同时,如何准确地从中检索出用户所需要的信息也已成为研究的热点,并开发出了很多信息检索系统,但传统的基于关键词的检索系统,其查准率和查全率都有待提高,鉴于此,文章提出了一种基于概念的信息检索系统模型,给出了它的理论模型和工作机制,其核心技术是自然语言处理,在其内部,查询和索引都是建立在语义层次上的,因而有较好的查全率与查准率。文章最后,对其发展方向作了展望,希望对致力于信息检索研究的同行们有所启发。
With sustaining development of Internet,it has become the biggest information database shared in the World.At present ,what how to find accurately information which are demanded has become a hot research domain,and many information retrieval systems has been developed,but for conventional keyword-based information retrieval systems ,their recall and precision need be improved.In view of this,a model of a concept-based information retrieval system is put forward in the paper,and its modal of theory and principle of work are clarified.Its nuclear technology is natural lan-guage processing,query and index are both based on semantic analysis,therefore,it is very efficient about recall and precision.At the end of the paper,its developing direction is forecasted.
出处
《计算机工程与应用》
CSCD
北大核心
2003年第36期173-175,181,共4页
Computer Engineering and Applications
基金
甘肃省科技攻关项目资助(编号:GS021-A52-054)
关键词
因特网
詹息资源
WEB
信息检索系统
人工智能
信息库
Information retrieval(IR),Natural language processing(NLP),Semantic analysis,Knowledge database,Index database,Automatic abstract
