Research On Crucial Technologies For Semantic Web Services

In recent years Web Services have become the predominant technologies with which applications on different machines communicate with each other over the Internet. Because Web Services rely only on open and standardized protocols, such as XML, SOAP and WSDL, they greatly enhance the interoperability of distributed applications. However, there are still some drawbacks that cannot be completely resolved with existing technologies. The fact that Web Services are described in terms of XML syntax without a well-defined semantics results in that computers can not process the interactions of Web Services automatically.This is where the Semantic Web comes to play. The Semantic Web is an effort to facilitate the web resources with semantic descriptions. It is not a separate Web but an extension of current one, in which information is given well-defined meaning, better enabling computers and people to work in cooperation. Ideally, Web Services and the Semantic Web should support each other but there are still some gaps. The Semantic Web Service is an attempt to bridge the gap between Web Services and the Semantic Web. The Semantic Web Services vision is to describe Web Services' capabilities in an unambiguous, computer-interpretable language and improve the quality and robustness of existing tasks. Semantic Web Services will greately enhance the automation tasks previously performed manually, including automatic discovery, invocation, composition, and interoperation of Web services.Automatic Web Services Discovery is an automated process to locate the Web Services that match the client's requirements appropriately. It is the basic issue of Semantic Web Services' research. This paper describes a model for flexible discovery of Semantic Web Services. The proposed approach has three main features compared with other approaches.Firstly, OWL-S and Reputation Ontology are used to describe the Web Services. As both a language and an ontology, OWL-S is widely used to add semantic annotations to Web Services and enable automatic discovery of Web Services. However, OWL-S is still evolving and has some shortcomings, especially that it can not describe the reputation of Web Services. In order to overcome this, the Reputation Ontology for Web Services is defined by the author in this paper. The proposed Reputation Ontology can describe whether a Web Service is trusty, as well as the Quality of Service and other things.Secondly, the matching algorithm is based on semantic similarity between the candidate services and client's request. The concept "Similarity" is used to describe how much the candidate services match the client's request constrains. Both OWL-S Profile Functional Similarity and Reputation Ontology Similartiy are calculated to measure the similarity between the candidate services and the client's request.Thirdly, the Reputation Manager of Web Services is introduced in theMatchmaker which is a software infrastructure that enables automatic discovery of Web Services. It is responsible for reputation management of Web Services. In the ideal concept model of Reputation Manager, there is a third-part Reputation Authority which is in charge of the measurement of Reputation for Web Services. But currently the third-part Reputation Authority does not exist, so a feedback mechanism is used by the Reputation Manager which collects the feedback from the clients and save them in a persistent database.