Improving Web Service Clustering through Post Filtering to Bootstrap the Service Discovery

Web service clustering is one of a very efficient approach to discover Web services efficiently. Current approaches use similarity-distance measurement methods such as string-based, corpus-based, knowledge-based and hybrid methods. These approaches have problems that include discovering semantic characteristics, loss of semantic information, shortage of high-quality ontologies and encoding fine-grained information. Thus, 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 the performance of clusters by rearranging services incorrectly clustered. Our approach uses context aware similarity method that learns domain context by machine learning to produce models of context for terms retrieved from the Web in the filtering process to calculate the service similarity. We applied post filtering approach to hybrid term similarity based clustering approach that we proposed in our previous work. Experimental results show that our post-filtering approach works efficiently.