Research On Oriented-Semantic Web Services Matching Technology

Services discovery and services composition are the applications of Web services. Services matchmaking is the key technology of sercices discovery. Because there are some disadvantages including low-recall and low-precise when using key word matchmaking algorithm, the research community brings semantics to Web Services. Semantic Web Services have their semantic descriptions and thus make it possible for services to know the content, function and attributes of each other. The Semantic Web should enable clients to locate, select, employ, compose, and monitor Web-based services automatically.The teaching system is designed in this dissertation and every part technology on this basis is analysed, and then the OWL-S matchmaker is designed in detail. The dissertation pay more attention to the key technologies.Firstly, the concepts of Semantic Web and its key technology are introduced. Secondly, following the principle of the ontology building, the ontology based systems of the semantic matching——elective ontology is established. Thirdly, the semantic description model is resigned.And then an effective service matching algorithm based on the Semantic Web is designed. The matching algorithm uses quantified similarity function to achieve the integration of performance and semantics and adjust the threshold value to achieve the fuzzy matching.At the last, a test matchmaking system has been proposed, and the part test results have been given. And the algorithm proposed in the paper has been proved working well through the analysis of the test results.Under the idea of the service function matching, getting rid of the registration services which don't belong to the requesting services through categories matching of services, and matchmaking various parameters of the service model. Users can request their own on the part of the different levels to set coefficients, decided the proportion of every part then use the similar function to calculate similarity degrees. Users can then set the desired threshold to choose all services, which accord with the matchmaking threshold, and eventually actualize the fuzzy matching purposes.