The research of lognormal probability distribution of software defects

摘要：

In recent years the distribution of software defects in software engineering was one of the most active domains, which plays an important role in the analysis of software quality, rationally allocating resources of software test, improving the efficiency of software test and making the balance of software cost. The paper selected representative software failure data to fit exponential distribution and lognormal distribution, and carried out KolmogorovSmirnov hypothesis testing. From a mathematical point of view, the paper proved the lognormal distribution more accurately reflected and described the software defect distribution, and provided evidence for the next step of the analysis of software defects, software testing.

关键词： software defects lognormal distribution Probability distribution exponential distribution Kolmogorov-Smirnov hypothesis testing

作者: Ouyang Xin Shuai Chun-Yan

作者单位: Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunmi Faculty of Electric Power Engineering Kunming University of Science and Technology Kunming, China

会议类型: 国际会议

会议名称: 2010 International Conference on Circuit and Signal Processing(2010年电路与信号处理国际会议 ICCSP 2010)

会议地点: 上海

会议语种:英文

页码: 400-403

在线出版日期: 2010-12-25（万方平台首次上网日期，不代表论文的发表时间）