Web Services Facile Discovery and Extraction by Latent Semantic Analysis

Service computing is a method for accessing data and web services in distributed environments. A web service is a software service that is identified by a URL and its public interfaces in the XML file. A significant issue in the service computing is to discover the appropriate web services with the user queries. Different researches have been done for discovering and classifying web services based on their WSDL file which they are non-semantic-based approaches. There are two semantic-based and non-semantic-based approaches. Existence a hybrid model of both approaches is essential for the web services accurate classifying and returning the most related web services to a user query. In this paper, a model is presented that discovers the latent concepts in the WSDL files and classifies the web services from two semantic and non-semantic perspectives. Then, it returns the most appropriate web services to the user query considering their latent concepts. To evaluate the proposed model, the various real web services in five different categories are used. The experimental results show that the presented model has a proper precision in classifying and returning the most related web services to user queries.

Web Services Facile Discovery and Extraction by Latent Semantic Analysis Keywords:

Service computing , Web service discovery , latent concepts , singular values decomposition (SVD)