基于改进多目标蜂群算法的Web服务组合优化方法

为解决Web服务组合优化方法中的组合多样性和服务质量的问题，在人工蜂群算法上提出改进，通过在算法中引入反向学习算子、精英引导策略和组合变异策略等操作，使得种群个体有针对性地进行更新，在保证服务组合质量的前提下，提高了服务组合的多样性.结果表明，所提算法具有良好的算法收敛性和均匀性，同时在为Web服务组合优化方面，也取得了较好的优化效果，提高了寻优精度、解的质量和收敛速度.

Web Service Composition Optimization Method Based on Improved Multi-objective Artificial Bee Colony Algorithm

To solve the problem of combinatorial diversity and service quality in Web service composition optimization methods， an improvement in artificial bee colony algorithm was proposed. Several methods such as reverse learning operator， elite guidance strategy， and combination mutation strategy were led into the algorithm， by which targeted information could be provided to update individuals. Furthermore， the diversity of service portfolios was enhanced on the premise of ensuring the quality of service portfolios. The experimental results indicated that the refined algorithm has fast convergence speed and good uniformity. Meanwhile， a better optimistic effect was also received for the optimization of Web service composition， and the search accuracy， solution quality and convergence speed were improved as well.