Font Size: a A A

Research Of Web Service Discovery Based On Semantics

Posted on:2012-05-20Degree:MasterType:Thesis
Country:ChinaCandidate:F YangFull Text:PDF
GTID:2178330335967084Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
With the rapid development of Internet, the number of Web services is rapidly growing. Finding out services those could meet users'requests from the huge number of services is the intension of Web service discovery. Description language and matching algorithm of service are the two kernel technologies to achieve Web service discovery. Currently, Service discovery methods are divided into syntax- level methods and semantic-level methods. Because the low performance of syntax- level methods, so ontology theory and semantic Web technologies are introduced in Web service by academia, and the service discovery methods based on semantics are proposed. There are two problems in current semantic service discovery methods: they could not implement service matching from the perspective of combining process and capability on the one hand, and do not use the pluralism relations of domain ontology concepts to match Web service on the other hand.To solve above problems, this paper proposes that, implementing service matching from two aspects of process-similarity and capability-similarity, meanwhile, the pluralism relations between domain ontology concepts will be used for service matching. The main job of this paper includes:1. Proceeding from the perspective of combing both process and capability, with the use of automata theory, Web services described by OWL-S are expressed as formalized processes. On this basis, a Web service capability matching algorithm based on process-similarity is proposed. This algorithm decides whether service request and Web service are process similar or not through process-similarity determineation, then the result obtained from process-similarity determination is used to match Web service capabilities.2. In the process of carrying out service capability matching, pluralism realtoins (including part/whole relation,synonymous relation and subclass relation) between ontology concepts are used to define a kind of semantic distance. Through this kind of semantic distance, the semantic similarity between service capability parameters corresponding to ontology concepts could be calculated.
Keywords/Search Tags:Web service discovery, OWL-S, automation, process-similarity, domain ontology, pluralism relations, semantic similarity, capability matching
PDF Full Text Request
Related items