A Rule-based Framework For Dynamic Semantic Web Service Composition

Web Service is one of the most important technologies of nowadays computer network. Some XML(Extensible Markup Language) based standards can be utilized for universal description, publish and data traffic of Web services, and this facilitates component sharing and interoperation of heterogeneous infrastructure platforms. Information systems of cooperative enterprises can communicate more easily over Internet by using Web service technology.A followed problem is how to efficiently select and integrate the inter-organizational services. Services provided on the Web should meet the user's requirements flexibly and credibly because of the instable and heterogeneous conditions of Internet. Therefore it is an important issue to provide a dynamic composition approach of services.It is faced with two essential problems to design a framework supporting the dynamic discovery, section and composition of semantically-described Web services via rule-based planning. Firstly, Web services are usually developed by different organizations that use different conceptual models for presenting services' characteristics. This requires utilization of relevant semantic information for matching and composition of Web services. Therefore, the framework describes Web services in a specification of semantic Web services called OWLS, and builds concept ontology repositories for every application domain. Secondly, there should be a mechanism to build workflow for service composition at the runtime. This is a hard problem and it is not entirely clear which techniques serve the problem best. One family of techniques that has been proposed for this task is AI (Artificial Intelligence) planning, and many research efforts tackling Web service composition problem via AI planning have been reported. The framework adopts rule-based planning to serve the composition problem by reasoning based on rulebases of every domain and semantic matching.Moreover, the adaptability and fault-contain ability of composition services is also an important issue. Therefore, framework divides different levels of granularity of loosely coupled workflow into abstract level and concrete level. As a result, the workflow could have a higher sharing level and be more flexible and robust.