Research On Heuristic WEB Services Composition Method Based On Planning Graph

In recent years, with the development of computer technology, Web services, as a distributed service providing solution, it has been widely recognized by the industry. But,often a single Web service is difficult to meet increasing user demand, therefore, how to reuse the existing fine-grained services, constructed complex services to meet the functional requirements and performance requirements become a hot research problem.With the combination of Semantic Web and Web services, semantic Web services have emerged, it makes the release of machine-understandable services possible and provides a good support for the service automatic composition technology. In this paper, semantic service for the study, proposed the heuristic Web services composition method based on Planning Graph. Service composition is essentially intelligent planning problems,and intelligent planning algorithm has been widely applied to the problem, but Service composition method based on artificial intelligence theory need pretreatment and formal transformation for services and tasks with high time complexity. If applied to the massive collection of services,search process of service composition and inference process will become very difficult. In this paper, Web services composition method based on Planning Graph for the study, in terms of planning solving efficiency and quality improvements proposed algorithm to improve, and they are summarized as follows:(1) through the study of service composition algorithm based on Graph Planning we found it add many of the services that are not related to the goal in the process of the Planning Graph expansion. In this paper, we proposed Graph Planning framework Web service composition algorithm based on status of distance, through state distance, evaluate the accessibility of the service, prune the Planning Graph and improve the efficiency.Furthermore, The approach is verified lower solution space consumption and higher efficiency without affecting the success rate by the experiment.(2) through the study of service composition algorithm based on graph planning, we found there is blindness in solution extraction stage and lead to a low quality programming solution. In this paper,we proposed Graph Planning framework QoS-based service composition algorithm which based on state quality evaluation function, and the function guides the plan solution extraction process, resulting in higher quality composite service.Finally, experiments show that the method has a higher solution quality and better computational efficiency.