软件测试数据自动生成算法的仿真研究  

Simulation Research on Automatically Generate Software Test Data Algorithm

作　　者：黄丽芬[1] 

机构地区：[1]广西工业职业技术学院建筑工程系,广西南宁530001

出　　处：《计算机仿真》2012年第10期245-247,382,共4页Computer Simulation

摘　　要：研究软件质量优化问题,传统遗传算法存在局部最优、收敛速度慢,使软件测试数据自动生成效率低。为提高软件测试数据生成效率,对传统遗传算法进行改进,提出一种遗传-蚁群算法的软件测试数据生成算法。针对测试数据自动生成的特点,充分发挥遗传算法的全局搜索和蚁群算法的局部搜索优势,提高了测试数据的生成能力。实验结果表明,遗传-蚁群算法提高了软件测试数据生成效率,是一种较为理想的软件测试数据生成算法。Testing data is the most crucial part in software testing software,and it is important for the software test automation degree to improve the automatic software test data generation method.Aiming at the defects of genetic algorithm and ant colony algorithm,a new software test data generation algorithm was proposed in this paper based on genetic and ant colony algorithm.Firstly,genetic algorithm which has the global searching ability was used to find the optimal solution,and then the optimal solution was converted into the initial pheromone of ant colony algorithm.Finally,the best test data were found by ant colony algorithm positive feedback mechanism quickly.The experimental results show that the proposed method improves the efficiency of software test data generation and has very important using value.

关 键 词：蚁群算法 遗传算法 软件测试 测试数据 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]