Semantic Web Service Discovery Based On QoS

Service-Oriented Architecture (SOA), for its open, intelligent and personalized features, meets growing network demands. How to choose the most users-requested service from a number of services has been many researchers critical concern. With the development of the Semantic Web, combining Web services with Semantic Web,Semantic Web Service Discovery came into being. However, the current Semantic Web services are generally used to establish Ontology to describe semantic information, and use logic reasoning for service matching. Although this way enhances the accuracy of matching of Web Service, it reduces the speed of matching. On the other hand, with the expanding of networks, network security and other service quality comes to be concerned. QoS (Quality of Service) protection for Web Service discovery becomes another research hotspot. Web Service discovery based on QoS can choose service further, and contain needs of availability, security and individual. Because of this, it can enhance the reliability of the process of Web Service discovery and meet the individual needs of users.In this thesis, for the first problem, it proposes to use WordNet ontology library for the matching of service parameters of Semantic Web service.WordNet ontology library uses the path and information of words in library to calculate the semantic distance between ontology. Then it uses the algorithm which the chapter provides to match input and output parameters. With the higher precision and recall rates and faster speed of choosing, we can obtain Semantic Web service discovery based on WordNet.Then, taking into account the importance of QoS guarantee, this thesis chooses performance, availability, network parameter and security, this four QoS properties as the QoS attributes of Web Service matching. Firstly, QoS ontology model is structured. It is divided into top-level, attribute level and measure level to richly describe the QoS information of Web service. Then, Based on this, a QoS-aware Web Service discovery model is established. It contains the matching of semantics and values of QoS to improve the accuracy of matching. Also it refers to advices of experts in the field, and considers the individual needs of users. At last, it sorts out the final selection of Web services for needs of users.Finally, this thesis evaluates two experiments to verify the solution. Compared to the traditional semantic Web service discovery based on OWL-S, the Web service discovery based on WordNet that this thesis proposes is feasible and relatively good. With presenting the process of QoS-aware Web service discovery, and compared with the traditional QoS matching, we found that the proposed method is more precise, and meets the individual needs of users better.