应用混沌果蝇算法的路径覆盖测试用例优化技术研究  

Optimization Techniques Research on Testing Data Through Path Coverage with Chaotic Fruit Fly Algorithm

作　　者：李龙澍[1] 郭紫梦 

机构地区：[1]安徽大学计算机科学与技术学院,合肥230601

出　　处：《小型微型计算机系统》2018年第2期362-366,共5页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(61402005)资助;安徽省自然科学基金项目(1508085MF127)资助

摘　　要：提出一种基于混沌果蝇的路径覆盖测试用例生成方法.鉴于果蝇优化算法与遗传算法等常用算法属于同一类型智能算法,且果蝇优化算法具有计算量小,复杂度低,寻优精度高等优点,故将果蝇优化算法运用到软件测试领域内,通过路径覆盖来实现测试数据的自动生成;并且针对果蝇优化算法表现出的易陷入局部最优问题,融入了一种新的混沌策略,对每次迭代过程中最优个体进行了改进,在保留优秀个体的同时,增加种群的多样性,优化全局搜索能力.最后,本文通过两组仿真实验,分别对比了在不同迭代次数下的覆盖率,及覆盖全部目标路径下的评价次数与运行时间,均取得较好的实验效果,验证了本文方法在路径覆盖测试领域内的有效性.This article proposes a method for generation of Software Testing Data Based on Path Coverage of chaotic fruit fly.Given that the fruit fly optimization algorithm,the genetic algorithm and other common algorithms belong to the same type of intelligence algorithm,besides,the fruit fly optimization algorithm has some characteristics such as small calculated quantity,less complication,and high optimization accuracy,I try to use the fruit fly optimization algorithm in the software testing area for the first time,and to realize testing datum′s automatic generation through path coverage;Moreover,for the issue that the fruit fly optimization algorithm is easily trapped in the local optimum,I take in a new chaotic strategy：make improvement to optimal individual in every iterative process,add population diversities and optimize the capability of global search while keeping excellent individual.In the end,through two simulation experiments,I respectively compare coverage under/with different iterations and evaluation times and runtime with all the target paths covered.As both experiments are well performed and good results are obtained,the validity of method in this article is testified in path coverage testing area.

关 键 词：果蝇优化算法 混沌策略 路径覆盖 测试用例 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]