《计算机应用研究》|Application Research of Computers

基于搜索的分层回归测试数据集扩增方法

Search-based hierarchical regression test suite augmentation method

免费全文下载 (已被下载 次)  
获取PDF全文
作者 王曙燕,高露,孙家泽
机构 西安邮电大学 计算机学院,西安 710121
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2019)07-035-2075-06
DOI 10.19734/j.issn.1001-3695.2018.05.0272
摘要 针对在回归测试中原有的测试数据集往往难以满足新版本软件的测试需求问题,提出一种基于搜索的分层回归测试数据集扩增方法,主要包含覆盖目标方法集获取模块和测试数据生成模块。首先对新版本程序进行抽象分析,提取出方法调用图,利用方法调用轨迹和已有测试数据建立方法覆盖信息,获取目标方法集,并通过计算贝叶斯条件概率对目标方法集进行优先选择;利用Hadamard矩阵设计正交种群,同时结合已有测试数据集进行种群初始化,采用文化基因算法对目标集中方法生成测试数据。该方法针对四个基准程序与随机法和遗传算法以及基于粒子群算法测试数据扩增方法相比较,测试数据的生成效率平均提高了95.2%、78.2%和50.5%,测试数据检错能力平均提高了47.9%、33.6%和18.2%,实验结果表明该方法更适合回归测试数据扩增。
关键词 方法调用图; 回归测试数据扩增; 正交种群; 文化基因算法
基金项目 陕西省工业攻关项目(2017GY-092)
西安邮电大学创新基金重点项目(CXJJ2017020)
本文URL http://www.arocmag.com/article/01-2019-07-035.html
英文标题 Search-based hierarchical regression test suite augmentation method
作者英文名 Wang Shuyan, Gao Lu, Sun Jiaze
机构英文名 School of Computer Science & Technology,Xi'an University of Posts & Telecommunications,Xi'an 710121,China
英文摘要 It is difficult for the original test data to meet the requirements of the new version of software testing in regression testing, thus this paper proposed a search-based hierarchical regression test suite augmentation method to solve the problem. The method mainly included obtaining the target function set module and the test data generation module. Firstly, it abstracted function call graph from the new program, and built function coverage information by using function traces and original test case set. Afterwards, it used Bayesian conditional probability to choose the target function set through calculating. Then it used Hadamard matrix to design the orthogonal population, and initialized population with the combination of the orthogonal population and existing test data set. Finally, it used the memetic algorithm to generate test data for target functions. This paper compared the proposed algorithm with the random algorithm based, the genetic algorithm based and particle swarm algorithm based test data augmentation methods on four benchmark programs. The results show that generation efficiency of the proposed method is improved on average by approximately 95.2%, 78.2% and 50.5% respectively, the error detection ability of the test case is improved on average by approximately 47.9%, 33.6% and 18.2% respectively. The experimental results show that the proposed method is more suitable for regression test suite augmentation.
英文关键词 function call graph; regression test suite augmentation; orthogonal population; memetic algorithm
参考文献 查看稿件参考文献
 
收稿日期 2018/5/16
修回日期 2018/7/13
页码 2075-2080
中图分类号 TP311.53
文献标志码 A