Context Aware Post-filtering for Web Service Clustering

Web service discovery is becoming a challenging and time consuming task due to large number of Web services available on the Internet. Organizing the Web services into functionally similar clusters is one of a very efficient approach for reducing the search space. However, similarity calculation methods that are used in current approaches such as string-based, corpus-based, knowledge-based and hybrid methods have problems that include discovering semantic characteristics, loss of semantic information, encoding fine-grained information and shortage of high-quality ontologies. Because of these issues, the approaches couldn´t identify the correct clusters for some services and placed them in wrong clusters. As a result of this, cluster performance is reduced. This paper proposes post-filtering approach to increase precision by rearranging services incorrectly clustered. Our approach uses context aware method that learns term similarity by machine learning under domain context. Experimental results show that our post-filtering approach works efficiently.