为提高Web组合优化算法的开发能力和运行结果的适应度、稳定性，以满足用户对Web服务的服务质量(Quality of Service，QoS)需求，提出了一种基于QoS模型和改良蜂群算法(modified Artificial Bee Colony，mABC)的Web服务组合优化方法。构建应用于Web服务组合优化问题的QoS顺序数学模型，使用基于混沌的对立学习方法，在进程运行的初始化阶段生成更好的初始群体，在蜂群算法的雇佣蜂阶段和围观蜂阶段使用新的相位搜索方程和围观搜索策略，有效地提高蜂群算法的探测能力和开发能力。将改良蜂群算法与人工蜂群算法(Artificial Bee Colony，ABC)、差分进化算法(Differential Evolution，DE)、改进灰狼优化算法(Modified Grey Wolf Optimizer，MGWO)、最优导向人工蜂群算法(Guided-best Artificial Bee Colony，GABC)和改进人工蜂群算法(Improved Artificial Bee Colony，IABC)进行了多次对比实验。实验结果表明，改良蜂群算法尽管在执行时间方面比其余算法都要略微长一些，但它在更为重要的适应度、稳定性方面都优于其余几种对比算法。
Web Service Composition Optimization Method Based on Modified Artificial Bee Colony
In order to improve the development capability of Web Combinatorial optimization algorithm and the adaptability and stability of the running results,and meet the user's demand for quality of service(QoS)of Web services,we propose a combinatorial optimization method of Web services based on QoS model and modified Artificial Bee Colony(mABC).The QoS sequential mathematical model applied to the combinatorial optimization problem of Web services is constructed.The chaos based opposite learning method is used to generate a better initial population in the initialization phase of the process operation.New phase search equations and spectator search strategies are used in the hire bee phase and spectator bee phase of the bee colony algorithm to effectively improve the de-tection and development capabilities of the bee colony algorithm,modified Artificial Bee Colony,Artificial Bee Colony(ABC),Differential Evolution(DE),Modified Grey Wolf Optimizer(MGWO),Guided-best Artificial Bee Colony(GABC),Improved Artificial Bee Colony(IABC)algorithms are compared and tested multiple times.The experiments show that although the modified ABC has slightly longer execution time than that of other algorithms,it outperforms other comparative algorithms in more important aspects of fitness and stability.
cloud computingWeb service compositionbee colony algorithmQoS attributechaos mapping
万方数据
维普
