摘要
由于传统的预测方法,在进行静态软件缺陷预测时只能进行局部预测,因此预测峰值低。基于此,本文提出基于大数据的静态软件缺陷预测方法研究。该方法在引入峰值识别机制的基础上,通过大数据技术对原始预测数据进行处理,计算静态软件缺陷预测映射误差;计算静态软件缺陷预测峰值,实现静态软件缺陷全局预测。实验结果表明,设计的预测方法预测峰值最高可达180.52gal,实验对照系统最高仅为121.45gal,设计的预测模型可以实现对静态软件缺陷的精准预测。
Due to the traditional prediction method,only partial prediction can be made when static software defect prediction is performed,so the prediction peak value is low.Therefore,a static software defect prediction method based on big data is proposed.Introduce the peak recognition mechanism,process the original prediction data through big data technology,calculate the static software defect prediction mapping error;calculate the static software defect prediction peak value,and realize the global prediction of the static software defect.The experimental results show that the predicted peak value of the designed prediction method can reach up to 180.52gal,and the experimental control system is only 121.45gal.The designed prediction model can realize accurate prediction of static software defects.
作者
张瑞
李学威
Zhang Rui;Li Xuewei(School of Information Engineering,Zhoukou Vocational and Technical College,Zhoukou Henan 466000,China)
出处
《信息与电脑》
2020年第19期43-44,共2页
Information & Computer
关键词
大数据
静态软件
缺陷预测
预测模型
全局预测
big data
static software
defect prediction
prediction model
global prediction
