应用遗传算法自动生成测试数据的实验分析  

Some Experiment Analysis of Using Generic Algorithm in Automaic Test Data Generation

作　　者：贾晓霞[1] 吴际[1] 金茂忠[1] 高仲仪[1] 

机构地区：[1]北京航空航天大学计算机学院,北京100083

出　　处：《小型微型计算机系统》2007年第3期520-525,共6页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(60373016)资助;国家"八六三"高技术项目(2004AA112030)资助.

摘　　要：将遗传算法应用于覆盖指定路径的测试数据生成已得到了广泛的研究.具体实现中,影响测试数据生成效率的因素很多,如先验知识,GA参数,路径复杂度等.本文在简要介绍应用GA生成测试数据的关键技术后,设计实验分析了影响测试数据生成的部分因素,并据此得出了一些结论.Applieations of Generie Algorithm to generate test data to cover the appointed path have been studied widely. In practice,there are a series of factors that affect the generation efficiency of the test data,such as transcendent knowledge,GA parameters ,path complexity etc. In this paper,key technologies of applying Generic Algorithm to test data generation are outlined at first. Then experiments are designed to analyze the factors that may affect the generation efficiency of the testing data. Conelusions are drawn according to the experiments and discussions. The results are useful for continuing research.

关 键 词：遗传算法 测试数据自动生成 软件测试 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]