Normalized similarity based semantic approach for discovery of web services

The practical challenges on the web are irrelevant and huge number of services returned by the UDDI and lack of standard mechanisms that helps in the discovery of desired web services. The utilization of the implicit semantic information from the service profiles can help the service consumers in selecting the most relevant services from a set of offered services. In this paper, an approach for web service discovery is proposed which uses a lexical semantic network constructed from the web snippets as a knowledge base for the calculation of semantic similarity between the service profiles. Our approach takes into account the text descriptions and involves mapping of service profiles to a category based dimension vector by using the notion of semantic similarity which is further merged with the IR based techniques of weight generation and is used for calculating the semantic degree of similarity between the services. We present results that we obtained by applying the approach on set of 106 OWL-S service profiles. Empirical evaluation shows that the proposed approach helps in better discovery of semantically similar and relevant services which are otherwise shown to be unrelated by the keyword based approaches.