论文部分内容阅读
针对半自动的Web服务组合模型,为了尽可能多地发现服务质量处在Pareto前端的服务组合供用户参考使用,提出了一种基于改进粒子群算法(MPSO)的Web服务组合推优方法.结合服务组合问题给出了粒子适应度评价函数以及群体多样性的计算模型.为了改善粒子群算法存在的早熟问题并且发现更多服务质量处在Pareto前端的组合服务,给出了受群体多样性指导的速度更新方法和惯性权重模型.针对指导粒子飞行的关键组合服务,给出了它们的寄存方法.最后通过实验从有效率和精确度及平衡性方面验证了基于MPSO的Web服务组合推优方法的有效性.
For the semi-automatic Web service composition model, in order to discover as many service quality Pareto front-end service portfolio as possible reference for users, this paper proposes a Web service combination push optimization method based on improved Particle Swarm Optimization (MPSO) In order to improve the premature problem of particle swarm optimization algorithm and find out more service quality in the Pareto front-end portfolio service, this paper gives a model that is guided by population diversity Speed update method and inertia weight model.Aiming at the key combination services which guide particle flight, their registration methods are given.Finally, experiments are carried out to validate the MPSO-based Web service combination push optimization method from the aspects of efficiency, accuracy and balance Effectiveness.