Research On Web Services Discovery Based On JXTA And OWL-S

Current Web Services Discovery technology is based on UDDI and WSDL and does not make good use of semantic information therefore to address the problem of locating Web Services through Web Services capabilities matching according to the demand of the requesters. On the other hand, centralized registries suffer from the traditional problems of centralized systems, namely performance bottlenecks and single points of failure. In addition, they may be more vulnerable to denial of service attacks. As problems arising on and on suggest that centralized registries should scale up to the needs of Web Services.Therefore they should develop towards the distributed architecture of high performance with semantic information.Semantic Web is a new direction of Web developing, and it concentrates on the semantic information of the Web. The Semantic Web Services add the semantic information to the definition of the Web Services, and help the Intelligent Agent discover and use the Web Services much more easily according to the semantic description of the Web services.P2P is a new peer-to-peer network architecture, it is a distributed computing mode which evolves from the traditional centralized system. The nodes in the purely scattered P2P system are all loosely-coupled over the web, and some of the nodes in the system are both service providers and clients, namely servants. Peers in the system can directly interact with each other without the interaction with the server, hence the P2P system can easily address the problems which exist in the centralized system mentioned above, it's an extensible sharing architecture with high performance.In this paper, we combine the Semantic Web, the P2P network and the Web Services Discovery together to solve the problems that exist in the traditional Web Services mentioned above, and present a Web Services Discovery model based on the JXTA and OWL-S which is used to describe the functionality of Web Services. And we also present a new semantic similarity algorithm by defining the intrinsic information value of a service description based on the inferencibility, and then we apply this similarity algorithm to a simulating test, finally we give the evaluation of the model.