Research On Semantic Web Services Combination Method Based On Graph And QoS

In recent years, with the Internet and electronic commerce development, moreand more service providers provide efficient services for the user in the form of Webservice. At present, a single Web service provide function is very limited, which can’tmeet the real life complex application requirements, so need to combination thesingle Web service. In addition, Web service which without semantic informationcan’t realize the automatic combination of Web services, but the semantic Webservices,which expand semantic for the Web service, make service automaticcombination becomes possible, so research on the semantic Web service combinationbecome a hot spot in the semantic Web service domain.At present, the mainly problems exist in the field of Semantic Web servicescombination are as followed:(1) Semantic Web services combination method needmeet the requirements of user more fast and efficient;(2) Semantic Web servicescombination method need to be improved in meet user’s requirements accuracy;(3)Semantic Web services combination method lack consider fast, flexible selectcombination services which have high service quality for users. To solve the problemmentioned above, the semantic Web services combination method based on the graphand the QoS(Quality of Service) is studied in this paper. At first, the relatedknowledge and technology involved in the semantic Web services combination isintroduced. The integration framework of semantic Web services combinationmethod based on the graph and the QoS is given next. Then the specific steps andrelated algorithm of semantic Web services combination based on the graph and theQoS is proposed by the direction of integration framework. At last, the experimentalis used to verify the feasibility of the method. The main work of this paper is asfollowed:(1) Put forward the method of semantic Web services combination based on thegraph.First,the method classified the semantic Web services according to the domainontology, it reduced the searching space and improve the query efficiency; Second, The method established the relevant rules library in each class, improved the timeefficiency in established the semantic Web combination graph; Third, the Web serviceabstract template is used in this paper. Users can modify the abstract template byusing the Jena reasoning machine and can generate combination path of abstractservices. It can better meet the user’s complex needs; Finally, this method establishedsemantic Web service combination graph according to the generated abstract servicescombination path and the user’s input and output request.(2) Propose the optimization method of semantic Web services combinationbased on QoS.First, user sets requirements about the weights preferences of servicequality attributes and the quality of service on domain attribute; second, the methodnormalized the conventional attribute and domain attribute of quality of service;Finally, this method choosed the optimum semantic Web service combination planbased on the heuristic algorithm of services combination and or graph, in the selectionprocess, calculated the attribute values according on the polymerization calculationrules of the quality of service. Compared to the existed only consider conventionalattribute method, this paper put forward the semantic Web service combinatorialoptimization method based on the quality of the service which calculated thecombination of conventional attribute and domain attribute, the final choice resultof the service quality of the combination service can more accord with user needs.(3) Design and realize prototype system of semantic Web services combinationbased on the graph and QoS using the related techniques and tools by the direction ofthe above theory. It contains four modules: the module to preprocess semantic Webservice, the module to generate abstract paths of semantic Web service, the module toproduce the graph of semantic Web services combination and the module to find theoptimal plan of semantic Web services combination. Verify the feasibility and validityof the method proposed in this paper through an experiment.