摘要
软件缺陷检测的主要目的是对程序模块中是否存在缺陷进行自动检测,以此有效促进软件的测试进程,使软件系统质量得到提高。针对传统软件缺陷预测模型的问题,提出在软件缺陷预测模型中使用粒子群优化BP算法。此模型使用粒子群优化算法对BP神经网络权值及阈值进行优化,通过交叉验证方法实现实验,并且同传统机器学习方法及BP神经网络等方法进行对比,实验结果表明提出的方法预测精准性比较高。
The main purpose of software defect detection is to automatically detect whether there are defects in the program module,which can promote the software testing process effectively and improve the quality of the software system.In allusion to the traditional software defect prediction model,the BP algorithm optimized by quantum particle swarm used in the software defect prediction model is proposed.This model uses particle swarm optimization algorithm to optimize the weights and thresholds of BP neural network.The experiment is carried out with cross-validation method.The method is compared with traditional machine learning method and BP neural network method.The experimental results show that the proposed method has high prediction accuracy.
作者
莫有印
MO Youyin(Shaoyang University,Shaoyang 422000,China;Chongqing Preschool Education College,Chongqing 404047,China)
出处
《现代电子技术》
北大核心
2019年第15期127-130,133,共5页
Modern Electronics Technique
关键词
软件缺陷
预测模型
量子粒子群
BP算法
交叉验证
预测精准性
software defect
prediction model
quantum particle swarm optimization
BP algorithm
cross-validation
prediction accuracy
