A SEMANTIC MODELING APPROACH TO AUTOMATIC SERVICES ANALYSIS AND COMPOSITION

摘要

E-services are services that make themselves available on the web and can be invoked over the Internet. These services can be composed together to form new services. With the great growth of the number of services on the web it is necessary to use an effective and efficient way to discover and compose them to meet the users' requirements. However, current web services description languages such as WSDL and OWL-S do not provide semantic support for automatic search and composition of services. Conceptual modeling has been considered to be an effective way to provide abstract descriptions for services that will greatly support automatic search and matches for services. In this paper, we propose a conceptual model for automatic search and matchmaking using four kinds of contextual information - metadata information, ontological information, functional information, and compositional information. Based on the conceptual model we discuss the contextual information for services and analyze their properties and interrelationships, as well as proposing a matchmaking algorithm to realize the contextual based automatic web service search and composition.