A Hybrid Web Service Selection Approach Based on Singular Vector Decomposition

With the number of registered Web services growing, Identifying desired Web service is crucial for Web users. Current keyword based service search are inefficient in two main aspects: poor scalability and lack of semantics. Firstly ,the users are overwhelmed by the huge number of irrelevant services returned. Secondly, the intentions of users and the semantics in Web services are ignored. We propose a hybrid approach of Web service selection that complements logic based reasoning with approximate matching. In particular, the large set of available Web services is first clustered into a set of smaller groups. Then a logic based reasoning with approximate matching based on Ontology is applied, which is followed by a syntactic matching method based on Singular Value Decomposition (SVD). Therefore, service matching is conducted within both syntactic-level and semantic-level.A set experimental results demonstrate that the proposed method outperforms several other alternative methods.