Web services discovery based on semantic similarity clustering

Efforts were made to match services using IOPE and nonfunctional attributes of the web service. Results of these service matching are not realistic as the matching´s done by analyzing description of the web service. In order to realize automatic Service matching accuracy there is a need to adopt web2.0 social participation about the web services into service matching modules to achieve desired result. In this paper we propose a new approach is proposed which use web2.0 social participation on web services and clustering methods. Initially services are clustered based on the semantic similarity score in co-relation with the functional semantic information of service specifications using wordnet2.1 (a Semantic Lexicon database) and tags descriptive of web2.0. Matching is done on clusters on the description generated within WSDL file to get the desired service. Both theoretical analysis and experimental results show that out approach is more efficient and increases recall ratio towards matching services. Relevance feedback from the user for a specific query are captured which helps in cleansing a query.