摘要： 针对语义Web服务发现效率低下的问题，提出一种基于图论聚类的服务发现方法。利用领域本体将Web服务形式化为领域概念的集合，根据概念间的关系计算服务间的相似度并构建用于标识服务的加权图，通过最小生成树算法对服务进行聚类，从而实现服务发现。实验结果证明，该方法有效减少了服务发现的匹配次数，提高了服务发现效率。

关键词: 领域本体, 聚类, 最小生成树, 语义网, 服务发现

Abstract: To solve the problem of efficiency of Web service discovery, this paper proposes a method for Web service discovery based on clustering. The method formal the Web service as the set of domain conception by using domain ontology and prevent the formula of service similarity calculation based on relationship of conception in ontology to build the weighted graph of Web service, and clusters the Web service by the utilization of Prim, thus realize serivce discovery. Experimental results show that the method can reduce the matching number and improve the efficiency of service discovery.

Key words: domain ontology, clustering, minimum spanning tree, semantic Web, service discovery

中图分类号: 

• TP391