Search space reduction approach for self-adaptive web service discovery in dynamic mobile environment

Abstract

The proliferation of functionally similar Mobile Web Service (MWS) result in huge search space, the discovery of MWS on such large space increases the response time and probability of discovering irrelevant MWS irrespective of the matchmaking algorithm. The existing research on MWS discovery mostly focused on applying coarse-grained search space reduction that fails to deal with cold-start and data sparsity challenges at the expense of large computing resources. The proposed search space reduction is achieved by subsuming k-means in the modified negative selection algorithm (M-NSA) to place the service in an appropriate category so that the matching is only performed on the MWS in the target category. The experimental results show significant improvement in terms of accuracy of the categorization which can improve the MWS discovery in in a dynamic mobile environment (DME).