Toward a More Accurate Web Service Selection Using Modified Interval DEA Models with Undesirable Outputs

Abstract

With the growing number of Web services on the internet, there is a challenge to select the best Web service which can offer more quality-of-service (QoS) values at the lowest price. Another challenge is the uncertainty of QoS values over time due to the unpredictable nature of the internet. In this paper, we modify the interval data envelopment analysis (DEA) models [Wang, Greatbanks and Yang (2005)] for QoS-aware Web service selection considering the uncertainty of QoS attributes in the presence of desirable and undesirable factors. We conduct a set of experiments using a synthesized dataset to show the capabilities of the proposed models. The experimental results show that the correlation between the proposed models and the interval DEA models is significant. Also, the proposed models provide almost robust results and represent more stable behavior than the interval DEA models against QoS variations. Finally, we demonstrate the usefulness of the proposed models for QoS-aware Web service composition. Experimental results indicate that the proposed models significantly improve the fitness of the resultant compositions when they filter out unsatisfactory candidate services for each abstract service in the preprocessing phase. These models help users to select the best possible cloud service considering the dynamic internet environment and they help service providers to improve their Web services in the market