制造资源本体的概念语义相似度研究  

Research on concept semantic similarity of manufacturing resource ontology

作　　者：丁博[1] 苗世迪[2] 

机构地区：[1]哈尔滨理工大学计算机科学与技术学院,哈尔滨150080 [2]哈尔滨理工大学软件学院,哈尔滨150080

出　　处：《计算机应用研究》2016年第1期28-31,共4页Application Research of Computers

基　　金：国家自然科学基金资助项目(60173055);黑龙江省自然科学基金资助项目(G201207)

摘　　要：为了实现制造资源本体之间的语义互操作,对本体中的概念进行语义相似性计算是进行此操作的关键技术之一。提出了一种计算概念语义相似度的新方法,将概念语义相似度分为主体相似度和附加相似度两部分。主体相似度综合考虑了概念自身的相似度、该概念的父概念和子概念间的相似度以及概念间的二元关系,同时,加入了概念属性相似度,属性携带了概念的大部分语义信息,计算属性相似度可以有效提高概念语义相似度的准确性;附加相似性是指通过本体中概念的层次结构对主体相似度进行语义补充,利用概念的深度对得到的概念语义相似度进行语义调整,有效地弥补了仅仅利用主体相似度计算概念语义相似度的不足。实例证明,该方法可以准确地计算同一本体和本体之间的概念语义相似度。In order to achieve semantic interoperability among manufacturing resources, the concept semantic similarity com- putation for ontology is one of the critical problems. This paper proposed a new method for calculating concept semantic simi- larity, which was divided into two parts： the main similarity and additional similarity. Main similarity synthetically considered the self-similarity of the concept, and the similarity of both parent concept and child concept, as well as the relationship be- tween every two concepts. In the meantime, it took into computing the similarity of concept attribute, for the attribute included a lot of semantic information. In the way, it greatly and effectively improved the accuracy of the concept semantic similarity. Additional similarity refered to the main semantic similarity complemented by ontology concept hierarchy, using the depth of concept to make semantic adjustments on the concept semantic similarity. Additional similarity could effectively compensate semantic similarity inadequate for the similarity calculation using only the main concept. Experimental result shows that this method can calculate correctly concept semantic similarity in an ontolo~ and between the ontologies.

关 键 词：本体 概念语义相似度 主体相似度 附加相似度 

分 类 号：TP391.43[自动化与计算机技术—计算机应用技术]