模糊关系在软件系统测试中的应用  

Application of Fuzzy Relations to the Software System Testing

作　　者：杨劲涛[1] 郭荷清[1] 

机构地区：[1]华南理工大学计算机科学与工程学院,广东广州510640

出　　处：《华南理工大学学报（自然科学版）》2004年第10期51-55,共5页Journal of South China University of Technology(Natural Science Edition)

基　　金：广州市重点科技攻关项目 (B2 - 10 9- 5 5 0 )

摘　　要：基于黑盒技术在分布式软件系统测试中的应用 ,提出了系统位态和预期位态的概念 .在执行系统测试时 ,测试用例对系统位态具有控制作用 ,系统位态与相应的预期位态则表现出模糊性 ,运用模糊蕴涵逻辑进行研究得到了如下结果 :测试用例对预期位态的作用关系是模糊控制关系 ,其控制程度取决于软件的质量 ;基于这种模糊控制关系和模糊推理规则 ,得到了被测软件通过测试的标准值 ,使测试通过的判定得以量化 ;文中还得出了量化的系统测试通过准则 ,并将其应用于自动化测试工具的改进 ,有效地提高了测试效率 ,降低了测试成本 .The concepts of system place-state (PS) and anticipative PS were firstly proposed based on the application of the black box technology to the system testing of distributed software. As the test case controls the system PS and a fuzzy occurs between the system PS and the anticipative PS, the fuzzy implication logic was next used to study the system testing. It was then concluded that the action relation of the test case to the anticipative PS was a fuzzy control relation and that the degree of control depended on the quality of software. The standard value of the tested software to pass the test was then obtained from the fuzzy control relation and fuzzy inference rules, thus quantifying the judgment for software to pass the test. Quantified pass-rules in the system testing were finally put forward and applied to improve the automatic testing tool, by which the efficiency of the test improved and the test cost was reduced.

关 键 词：测试用例 软件系统 分布式软件 模糊控制 系统测试 自动化测试 模糊推理规则 预期 控制程度 执行系统 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论] F270[自动化与计算机技术—计算机科学与技术]