主客观权重遗传算法在服务选择中的研究  

Research on genetic algorithms based on subjective and objective weight in services selection

作　　者：吴明礼[1] 魏瑞珍 WU Ming-li;WEI Rui-zhen(School of Information,North China University of Technology,Beijing 100144,China)

机构地区：[1]北方工业大学信息学院,北京100144

出　　处：《计算机工程与设计》2020年第3期729-734,共6页Computer Engineering and Design

摘　　要：为解决Web服务选择中服务质量(quality of service,QoS)属性的权重只注重用户偏好以及选择算法收敛速度慢的问题,提出一种基于主客观权重的遗传算法。采用键值对Web服务进行编码,适应度函数计算中融合主客观权重,引入小生境思想和集中因子对交叉算子和变异算子进行改进,对Web服务的选择模型进行求解。实验结果表明,所提遗传算法在收敛速度方面较其它算法更快,最终的服务选择结果符合开发人员需求的同时性能也更好。To solve the problems that the weight of quality of service(QoS)attribute focuses on user preference and the convergence speed of selection algorithm is low in Web service selection,a genetic algorithm based on subjective and objective weight was proposed.In the selection model of Web services,key pairs were used to code the Web services,the subjective and objective weights were fused in the fitness function,and the crossover operator and mutation operator were improved by introducing niche idea and concentration factor.Experimental results show that the genetic algorithm based on subjective and objective weights has higher convergence speed than other algorithms,and the final service selection results meet the needs of developers with better performance.

关 键 词：WEB服务 服务质量 主客观权重 遗传算法 服务选择 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]