Research On Web Service Matchmaking And Selecting Method Based On Semantics

The web services users are very difficult to get the best services unless there is a very effective method about Web services matching and selecting, so the study of Web services matchmaking and selecting is very useful. This thesis focuses on this issue.Firstly, the research background of the thesis is briefly introduced, after which the state of the art of the relevant technology about Web services matching and selecting is elaborated.Secondly, aiming at lacking of considering the service behavior constraints and the deficiency of the traditional reason matching whose result can not tell the difference between the same degree in current matching algorithm of semantic web service, a new matching approach of semantic web service with service behavior constraints based on the reasoning and the computation of semantic similarity is proposed. It first checks the compatibility of the service behavior constraints, and then reasoning and computing the semantic similarity between the services and get the matching degrees.Compared to OWL-S/UDDI(Web Ontology Language, Universal Description Discovery Intergration), the results shows that this approach could increase recall and precision ratio greatly.Finally, aiming at lacking considering the credibility of the QoS data in Web services selecting methods and based on the fact of many false services in the network, a Web services selecting method considered the credibility of the QoS data is proposed. It first computes the credibility of the QoS data using the method of error analysis, and then computing the weight coefficient of the QoS, the credibility of the QoS data and the matching degree, finally computing the overall evaluation. Theoretical and example analysis confirmed that this method can effectively weaken the influence of the malicious services on the process of service selecting, thus it can strengthen the accuracy of the service selection.