Rank Aggregation for QoS-aware Web Service Selection and Composition

Service Composition is an essential part of Service<br> Oriented Architectures (SOA). The number of available alternatives<br> for web services offering similar functional characteristics<br> opens opportunities to maximise compositions based on criteria<br> embedded into the Quality of Service (QoS). These non-functional<br> criteria may be numerous and diverse, and arise from various<br> constraints, making composition an optimal information fusion<br> problem. A number of solutions have been offered to resolve the<br> selection problem for web service composition, essentially relying<br> on exploiting and combining absolute performance values for<br> measuring multi-dimensional QoS.<br> In this paper, we investigate the use of Rank Aggregation<br> methods for performing adequate information fusion for web<br> service selection in view of composition. Our goal is to take<br> advantage of these methods working on ranks rather than on the<br> direct values for inferring global QoS. We formalise our approach<br> and demonstrate its validity via an empirical study. We further<br> evaluate our approach on publicly available real data gathered<br> from actual web service characteristics.