Research On Semantic WEB Service Composition Framework

As rapidly as web technology is developing, web service as a distributed application technology is being more and more emphasized. As higher integration level and service performance is required, the attention of industry and academic research is transiting from individual web service to composite web services. Semantic is a key technology in solving service composition problems. Following such a trend, the research this thesis involves is about web service composition based on semantic.Based on a thorough investigation of web service composition related technology and research activities, this thesis refers to and semantically enhances current industry standards, and implemented them through the service composition lifecycle, forms an integrated, highly realistic, compatible, and extendable service composition framework, based on which in depth research on semantic annotation of service description and service matching and selection algorithms are carried on.In the research on two key modules of the framework, service description and semantic annotation module implements a mapping approach among service description and ontology. Matching degrees are evaluated and ontology concepts are selected for semantic extension. Through experiment, the performance and application areas of the algorithm are evaluated and results are analyzed. During the design of models and algorithms in service matching and selection, formalized description of semantic service is defined, template matching and service compatibility degree algorithms are developed. A process structure based reduction algorithm is specifically designed for service selection, which effectively reduces the computing complexity for process overall matching degree and for selecting services among large amount of candidates.