关键词: Web服务选择, 不确定QoS, 概率区间, 约束优化, 遗传算法

Abstract: To deal with the dynamic and stochastic characteristics of Web service in open environment, a Web service selection algorithm based on uncertain Quality of Service (QoS) was proposed. The uncertainty of QoS was described by probability interval, including confidence upper bound, lower bound and the mean value. By using the difference between the mean value and the two bounds, the influence of dynamic environment on Web service was calculated. Thus, the selection of optimum Composite Web Service (CWS) could be evaluated with the maximum weighted sum of mean and the minimum difference. The Web service selection problem under dynamic environment was transformed into single-target combinatorial optimization problem with uncertain QoS constraints. Through the Genetic Algorithm (GA) based on the community partition, the algorithm convergence was speeded and the service composition which satisfied the optimum conditions was obtained. The experimental results showed that the proposed algorithm could overcome the effect of uncertain factors on Web service selection, and the selection efficiency was improved as well as the selection was ensured.

Key words: Web service selection, uncertain quality of service, probability interval, constraint optimization, genetic algorithms