Research On Semantic-based Web Services Discovery Technology

With the proliferation of Web Services, it is becoming increasingly important to find a Web Service that will satisfy our requirements. UDDI is emerging as the standard registry for Web services and it is proposed as the main tool for Web service discovery. However, the only discovery mechanism provided by UDDI is keyword search on the names and the classification with low precision, recall and automatization degree. The object of the Semantic Web Service discovery technology is achieving the automatic service discovery. It can enhance the quality of service matching by improving the machine's understandability of service description with ontology and logical reasoning in service discovery phase. The research hotspot of the domain is how to realizing automatic service discovery. This thesis particularly focuses on the research on Semantic web services discovery technology, it specially studies the description of Web Services and Service Matching.The object of the research is improving the automatization degree of the Service discovery. Web Services and Semantic Web technologies are studied. First, OWL-S is analyzd, and an Ontology Quality of Service Description Language for Web services is defined to make up the limitation of Quality of Service description. Then, the algorithm of Grade matching is analyzed and improved by many means. A matching grade is added to the algorithm by calculating the Binary Relation, Similarity computation is added by combining with the Network Distance Models, a method of matching result standardization is proposed to operate the concept matching result, the service described by OWL-S and OQDL-S is supported by the algorithm. At last, a Matchmaker is implemented, it's a validating to validity of the methods which proposed by this thesis. It adds a Semantic Matchmaker and a QoS Matchmaker to UDDI. A mechanism of Web Services validity and reputation feedback and management is proposed in the QoS Matchmaker. A experiment is designed to the algorithm, the experiment result shows that the improved algorithm has higher precision and recall Top-k precision,it can support the automation of service discovery well.