Research On Web Service Discovery Based On Semantic

Web Service Discovery is an important component of Web service architecture. Discovering Web services to meet the needs of users is an important prerequisite to achieve Web service reuse and composition. With the rapid development of Web service and the sharp increase in number, effective Web Service Discovery has become a challenging issue, particularly in the automatic service composition and interoperating process of complex service. Therefore, it is one of the key issues to be resolved at this stage that how to discover services meeting users from a large number of Web services in a rapid, accurate and efficient way.In order to resolve the existing problems of Web Service Discovery, the semantic-based Web Service Discovery integrating with Semantic Web technology is proposed in the paper. In order to achieve the effective matching between service requests and registered services, a Web service description model and a corresponding description model of the service request are given, describing functional and non- functional attributions of services more comprehensively.At present, on account of the diversity and one-sidedness of the existing concept similarity algorithms, an integrated semantic similarity algorithm is put forward to take the impact factors on the similarity between ontology concepts into account in a more comprehensive way. On this basis, the function matching between services is researched in the paper to improve the recall and precision of Web Service Discovery effectively.In the meantime, a semantic-based Web service discovery model and a semantic-based Web service discovery mechanism with clustering, are presented to enhance the efficiency of Web service discovery and the availability of the found services. By using the clustering algorithm, the registered Web services with the certain similarity are grouped together. While service discovery, firstly, the similarity between the description of a service request and the service clustering is calculated; then, non-clustering-related services are filtered according to the similarity, which can improve the efficiency of the service discovery through reducing the processing time. And finally, the combination of the service quality is used to choose the optimal services. In the paper, Web service description, service matching algorithm, Web service discovery model and Web service discovery mechanism are studied to improve the performance of Web service discovery: the recall, the precision and the efficiency.