基于语义加权距离的语义相似度改进算法  

An Improved Algorithm for Semantic Similarity Based on Weighted Semantic Distance

作　　者：徐桂臣[1] 叶枫[1] 

机构地区：[1]浙江工业大学经济贸易管理学院,杭州310023

出　　处：《情报杂志》2012年第2期119-123,共5页Journal of Intelligence

摘　　要：在信息检索领域,概念的语义相似度计算起着重要的作用。针对现有研究文献中语义相似度计算中的若干问题,本文在语义加权距离算法的基础上讨论了多继承问题和概念属性问题。对于多继承问题,探讨了概念之间通过父概念和子概念进行连接所得到的路径长度相等的情况;对于概念属性问题,本文将对象类型属性和数据类型属性区别对待,提出了数据类型属性的相似度算法。最后,通过计算一个实例中概念节点的相似度,证明了本文算法的有效性。The concept of semantic similarity calculation plays an important role in information retrieval. To solve the current problems of semantic similarity calculating in the related literature, this paper discusses the problem of multiple inheritance and concept attributes on the basis of the algorithm of weighted semantic distances. As for multiple inheritance problem, the paper only investigates the case when the length of path is equal through parent concept＇s link and sub-concept＇s link; as for concept attributes problem, the paper makes a difference between object property and data property and proposes the semantic similarity algorithm of data property. Finally, the paper proves the effectiveness of the proposed algorithm by calculating similarity of the concepts in the example.

关 键 词：本体 语义相似度算法 语义加权距离 多继承 概念属性 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]