A QoS-oriented Optimization Model for Web Service Group

With the ever increasing of Web services in a Web service group, in which Web services offer the same functionality with different QoS parameters, Web service selection in a Web service group is an ongoing research topic. In this paper, we propose a QoS-oriented optimization model for Web service selection from a Web service group. The proposed model builds up a lightweight QoS model firstly, which defines functionality satisfaction degree, performance satisfaction degree, cost satisfaction degree and trust satisfaction degree as the QoS parameters of a Web service. Secondly, the Web service optimization model is described formally. Based on the described optimization model, the QoS vectors are clustered by hierarchical merging clustering approach. The clustering process is determined by the maximum similar degree parameter. QoS predication and QoS parameter predications are produced after the clustering is finished. Finally, the optimization is carried out by two searching modes: the clusters are optimized and the optimized Web services are selected from the optimized clusters by using of decision tree, QoS predication and QoS parameter predications. It is proved by our simulation cases that the proposed model is effective and can be used for Web service selection from its group.