基于角色及语义相似性的语义Web服务发现

摘要

在Web服务发现过程中由于容易忽视用户角色的作用,往往导致难以发现到合适的Web服务.在传统本体语义Web服务发现的基础上,结合RGPS (Role-Goal-Process-Service)元模型框架,文中提出了将用户的角色加入服务发现本体中,使用KM算法计算采用二分图最优匹配获得的词汇语义间的相似度,同时结合使用“WordNet”进一步计算词汇语义的相似性,进而提高查找的准确率,最终找到能够满足不同用户需求的语义Web服务.实验结果表明服务发现方法在查准率和查全率方面相比传统基于关键词的方法有一定的改善.%It's hard to find the fittest Web Service because of neglecting the role of users in the course of Web service discovery. On the basis of traditional ontology semantic Web service discovery, combining with RGPS meta model framework, the role of users is added into the ontology of OWL-S, and KM algorithm is taken to calculate the semantic similarity of vocabularies by bipartite graph optimal match. "WordNet" is utilized to further improve the semantic similarities of vocabularies and the search accuracy so that the Web services can meet the needs of different users. Experiments show that the proposed method can achieve better results than traditional keywords method in precision ratio and recall ratio.