一种面向语义的信息检索方法  

A Semantics-Based Information Retrieval Method

作　　者：张明宝[1] 马静[1] 

机构地区：[1]南京航空航天大学经济与管理学院信息管理与电子商务系,南京210016

出　　处：《情报学报》2009年第4期509-515,共7页Journal of the China Society for Scientific and Technical Information

基　　金：南京航空航天大学引进人才基金（1009-234039）

摘　　要：传统的信息检索技术忽视了语义对检索过程的影响,这是造成查准率不高的一个重要原因。论文提出了一种面向语义的信息检索方法,该方法强调使用基于知网的语义处理技术实现对用户查询需求和目标文档的语义标注,使用基于知网的词汇链技术实现对文档特征词汇的过滤。一方面可以实现语义级别的检索匹配,另一方面可以降低大量无关词对检索结果的干扰。论文描述了一个实现该方法的信息检索系统SOIRS,并且利用该系统与传统检索系统做了对比实验。实验结果表明面向语义的信息检索方法在查准率方面要明显优于传统信息检索方法。The important effect of semantics processing to conventional information retrieval technology is neglected,which may be one of the main reasons for dissatisfactory veracity rate of information retrieval.This paper presents one kind of semantics-based information retrieval method.The method stresses to utilize two kinds of technology based on HowNet,one is using semantics processing technology to label documents and user＇s demands,and the other is using word chains technology to filtrate stamp word of documents.And two aspects of advantages may be brought,one is achieving matching of information retrieval at semantic level,the other is decreasing the disturbing of lots of irrelevant word to information retrieval.One prototype system, named as SOIRS and directed by the method,is introduced.And finally one experiment compared to traditional search engine is displayed whose results prove that semantics-based information retrieval method has a higher veracity rate than conventional method.

关 键 词：信息检索 语义处理 词汇链 

分 类 号：G354[文化科学—情报学] F832[经济管理—金融学]