基于缺陷分类和缺陷预测的软件缺陷预防  

Software defect prevention based on defect classification and defect prediction

作　　者：雷挺[1] 

机构地区：[1]中国电子科技集团公司第十五研究所,北京100083

出　　处：《计算机工程与设计》2013年第1期215-220,共6页Computer Engineering and Design

摘　　要：为降低软件缺陷率,对现有的缺陷预测模型进行了优化,同时引入正交缺陷分类方法,并对该方法加以改进,使其能够支持缺陷的原因分析,将缺陷预测与改进的正交缺陷分类方法结合起来,形成一套软件缺陷预防流程并应用在实际项目中。实验结果表明,该成果可以在软件生命周期的各个阶段有效预防缺陷,最大限度地提高软件质量。In order to drive down error rates in software, the multiple AODE defect prediction model has been optimized as well as the ODC method which can assist defect cause analysis has been introduced and improved. Finally, a process of software defect prevention is put forward through a combination of the defect prediction with the improved ODC method. Practice has proved that this technological process can be applied in all stages of software projects, prevent software defects effectively, and improve software quality in the greatest degree.

关 键 词：改进的正交缺陷分类 缺陷预测 缺陷预防 缺陷数据集 缺陷原因分析 

分 类 号：TP311.5[自动化与计算机技术—计算机软件与理论]