An Empirical Approach for Semantic Web Services Discovery

Component retrieval/discovery is a well-established research direction in Software Engineering. With the surge of Service-Oriented Architecture (SOA), service discovery has become increasingly crucial. However, the public UDDI Business Registry - the primary service discovery mechanism over the Internet - has been shut down permanently since 2006. Moreover, keyword-based service discovery is insufficient in coping with complex discovery requirements posed by modern software developers. In this paper, we propose an empirical semantic- based Web service discovery approach. It provides an automatic Web service discovery mechanism that can locate relevant Web services based on concepts rather than keywords. The major contribution of this paper is three fold. First we articulate three requirements that software developers often raise during the component/service development and discovery process. Next, we propose the application of Latent Semantic Analysis into the area of Web services discovery. To our best knowledge, little work has been done in this area which leverages concept-based Information Retrieval models in service discovery. Last, we provide a proof-of-concept system prototype that can suffice three specific requirements of semantic service discovery.