基于主题图的本体信息检索模型研究  

Research on topic maps-based ontology information retrieval model

作　　者：李清茂[1] 杨兴江[1] 周相兵[1] 马洪江[1] 

机构地区：[1]阿坝师范高等专科学校计算机科学技术系,四川郫县611741

出　　处：《计算机应用》2010年第1期240-242,248,共4页journal of Computer Applications

基　　金：四川省教育厅自然科学基金资助项目(07ZC002)

摘　　要：针对本体在定义领域概念时具有规范性、明确性和可共享性等特点,结合主题图对文档资源组织方式具有语义可导航性,提出了一种基于主题图的本体信息检索模型,并给出了模型的形式化定义。选择旅游领域作为研究对象,定义了旅游本体和旅游文档资源主题图,分析了在信息检索模型中利用本体来规范用户自然语言查询输入,识别用户检索意图和扩展查询语义方面的作用,并展示了主题图在语义导航和用户相关度排序方面的价值。最后通过实验表明基于主题图的本体信息检索模型较传统的检索系统有较好的性能。Ontology is normative, explicit and reusable when defining the domain concept, so it can be combined with topic maps to organize information resource for semantic navigation. An information retrieval model based on topic maps and ontology was proposed and defined formally. Firstly it specified a domain of tourism document. Secondly it defined the ontology and topic maps of tourism document in order to normalize query that user directly input in natural language, and identified the user＇s real meaning of search. Thus, it can expand user＇ semantic search. Therefore analyzed the effect of the ontology was analyzed, and a valuable function of semantic navigation and sorting the retrieval result correlated with user＇s query was shown. Finally, the experimental result shows that the topic map-based ontology information retrieval model can perform better than the traditional model.

关 键 词：本体 主题图 信息检索模型 语义扩展 

分 类 号：TP319[自动化与计算机技术—计算机软件与理论]