Popularity Based Web Service Search

Web service discovery is a main challenge despite the enhanced proposed methods based on information retrieval techniques (word sense disambiguation, stemming, etc.), domain knowledge and ontology. Unfortunately, the proposed approaches are, however, complex in practice. Despite the addition of extra information to WSDL documents, discovering a required web service looks like finding a needle in a haystack. Thus, we propose in this paper a seamless way to discover a more appropriate web service independently of its type (simple, composed or semantic). In fact, we consider the usage of the service and more particularly, users' requirements, and we invoke it using a generated GUI. This approach concerning a Web service search engine based on popularity is implemented and validated by performing an experiment.