DefinePK

Abstract

Web services are software entities that have a well defined interface and perform a specific task. A number of web services are composed together to fulfill the customer requirements because single web service not always fulfill the customer requirements. It can be grouped based on different similarity measures. The domain-ontology-based PSO-inspired Balanced Iterative Reducing and Clustering using Hierarchies (BIRCH) clustering and Improved Bipartite graph (IBG), determines the relevancy between web services based on concept distance similarity measure. Based on this similarity measure the web services are clustered and ranked. The major drawback of this method is, there is no single similarity measure is optimal for determining the relevancy between web services. So in this proposed work, an efficient clustering method is proposed that uses different matchers with different similarity measures to calculate the relevancy between web services. It enhances the improved bipartite graph based web service matching. But the different matcher may return different clustering results, so a majority voting algorithm called as Boyer-Moore majority vote algorithm is introduced to finalize the clusters. After clustering the web services, the web services are ranked based on QoS based Fuzzy clustering. Experimental results show that the proposed method the efficiency of web service composition in terms of accuracy, runtime, recall and precision.