Research On Ontology Based Semantic Description For Web Services QoS And Service Selection

The QoS (Quality of Service) of Web services is used to describe the ability of services to satisfy the requirement of the requesters. With the rapidly increasing number of Web services which share similar functionality, how to select services that meet the demand of requesters with celerity and accuracy becomes a very important problem to be solved.Based on the analysis of the requirement of QoS description model, the concepts for Web service QoS and the relations among them are studied. Following the work above, a OoS ontology that owns a good extensibility and powerful semantic description capability is built by extending the semantic Web service description language OWL-S, realizing the semantic description of QoS information. Then, the expressiveness of the QoS ontology is verified by an example.With the semantic description of QoS information, the QoS metrics are matched based on semantics according to the different QoS metric ontology instances given by the service provider and requester. For the two types of quantitative and qualitative QoS metric, the corresponding algorithms is presented, through which the services without the QoS metrics needed by the requester is filtered. Furthermore, the match of the values for the quantitative QoS metrics is processed. A service selection algorithm considering the matching degree of the values of the QoS metrics, which also extends the matching of the interval values, is proposed to improve the traditional selection algorithm. As a result, the personalized needs of customers could be sufficiently met. At last, an experiment designed to compare the selection effect is implemented, and the result proves that the algorithm could select the best services on the basis of satisfying the demand of the requester and takes advantage in realizing the personalized service selection.