Semantically structured service community discovery: based on relationship and functionality

Abstract

As a key enabler to achieve the full potential of service computing, service community discovery is employed to support web service intelligent application effectively. The proposed framework applies a semantically structured approach to the web service community modelling for service community discovery based on probabilistic graphical model and probabilistic topic model. The approach to the problem is to identify web services of a service community based on both the semantics of the functional properties of web services and the graph structure of the inherent semantic association, dependency among services, which not only leverages the semantic functionality property but also exploits the semantic association relationship between services and operations. The experimental results and analysis shows that the proposed approach is efficient, feasible and practical. The result of the present work implied that the service community can be discovered by the approach and also has the full semantic interpretation.