摘要
软件测试是确保软件产品质量的有效技术手段,其根本目的是发现隐藏在软件中的缺陷,并通过对其的修复尽可能减少遗留在系统中的缺陷数量,以提升软件的质量。随着缺陷数据的不断积累,面对庞大、甚至海量的缺陷信息,已无法通过人工方式进行缺陷分析。基于此,作者对缺陷分类和数据挖掘技术开展研究,总结测试工程实践中的缺陷特点,提出改进的正交缺陷分类模型;结合数据挖掘中的关联规则挖掘算法,提出缺陷关联分析模型。并对上述模型进行应用说明,帮助软件技术人员定位和解决缺陷,提供软件测试缺陷分析的辅助手段。
Software testing is an important means to ensure the quality of software products. The target of software testing is to find defects and repair them as much as possible, finally to improve the quality of software. With the continuous accumulation of defect data, it is impossible to perform defect management and analysis artificial way, automaticmethod is necessary for reasonable and effective use of defect information. Base on the development of softwaredefect management and data mining technology, the author summarized the characteristics of defects in software testing,and puts forward the improved orthogonal defect classification model and defect association analysis model. The application of the above model is explained, which can help the technicians to locate and solve the defects, and to provide the auxiliary means for software testing defect analysis.
作者
颜乐鸣
YAN Yue-ming(The 7th Research Institute of China Electronics Technology Group Corporation, Guangzhou 510310, China)
出处
《软件》
2017年第1期70-76,共7页
Software
关键词
软件测试
软件缺陷分析
正交缺陷分类
关联规则挖掘
Software Testing
Software Defect Analysis
Orthogonal Defects Classification
Association Rules Mining
