本体映射中的概念相似度计算  

Computation of Conceptual Similarity in Ontology Mapping

作　　者：郑诚[1] 秦多荣[1] 

机构地区：[1]安徽大学计算智能与信号处理教育部重点实验室,安徽合肥230039

出　　处：《计算机技术与发展》2008年第11期125-127,137,共4页Computer Technology and Development

基　　金：国家自然科学基金资助项目(60475017);安徽省高等学校自然科学研究项目(2006kj055B)

摘　　要：本体是概念、属性和关系的集合,本体映射是解决本体异构的最好方法。文中针对目前本体映射过程中概念相似度计算存在的问题,提出一种综合的相似度计算方法。先根据本体中两个概念名称的相似性,选出最相关的概念,减少相似度的计算,然后分别基于概念的属性、实例和关系来计算概念相似度,并进行综合得到概念相似度。在计算属性相似度时,先通过计算属性的信息增益来确定各个属性的优先级,最后只选取几个信息增益大的属性进行相似度的计算,从而减小计算量。Ontology is usually viewed as the sets of concepts, attributes and relations, ontology mapping is the best way to .solve ontology heterogeneity. To aim at the current problems of the computation of concept similariW, puts forward an improved approach. Firstly, the most related concepts are filtered out according to the similarity between two concept names so as to reduce the similarity computation, and then an integrated approach of based- concept attribute, besed - concept instance, based - concept relation is designed to compute the concept similarity. When calculating attribute similarity,determine the priority of various attributes by computing their entropies, then only compute the attribute similarity through several attributes with relative large entropy values to reduce the amount of computation.

关 键 词：本体映射 概念相似度 属性相似度 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]