Semantic Web Service Composition Based On Qos

In recent years, with the rapid development of the software industry, service-oriented computing (SOC) and service-oriented architecture (SOA) has become the latest developments in the direction of software development and distributed computing, which is done by connecting a business, building applications and business processes to complete particular tasks. Owning to its loosely coupled, SOA can help the companies to add new services or compose existing services more quickly, cheaply and flexibility. It can provide services in many different ways to increase the competitiveness of enterprises. In the face of fierce market competition, to quickly build on-demand business process applications and Web service composition is becoming the core of its technology. With the rapid development of Internet and c-commerce technologies, more and more web service compositions occur in Internet environment. Service-oriented architecture provides a flexible framework for service composition. Web service composition enables seamless and dynamic integration of business applications on the web, which has recently taken a central stage as an emerging research area. The key research issues of web service composition include the web service description, service discovery, service selection, web service composition, composition verification, service evaluation, monitoring and management. However, a large proportion of today's data on the Web are understandable only to humans or custom developed applications. Thus, adding semantic web to web services technology can make the service automatic matching, discovery and composition possible. How to dynamically select web service that can best meet the requirements of consumers has also become a serious problem.Firstly, QoS-based web service dynamic composition framework is presented. Based on the full reference on relevant research, this paper proposes QoS-based service composition framework supporting semantic matching. It includes five modules: request agent, composition agent, selection agent, executor engine and QoS evaluators and adaptation. Then the role and main functions of each module is introduced.Secondly, considering the service input, output, precondition and effect, the relationship between services is measured by calculating similarity of concepts. Then service matching method sort services by the similarity results. In the service discovery process, it takes both service precondition and effect into account and returns sequence of services to meet user's function requirements.Thirdly, we research on service composition method which merge particle swarm optimization and simulated annealing operations to solve the QoS-based service selection problem. The model about the user preference of QoS considers semantic ambiguity, which can reflect the user's QoS preferences and provide a clear basis for the service composition algorithm more accurately. The algorithm makes full use of user's preference and user's evaluation information to exactly meet customer nonfunctional requests. At last, experiment results have shown the feasibility and effectiveness of the proposed method.