摘要
提出基于改进的粒子群优化支持向量机方法(PSO-ISVM)的测控软件缺陷预测方法。通过引入代价惩罚系数,定义粒子群优化算法中的适应度函数,利用最小化适应度函数值作为优化目标,排除大量的冗余干扰信息,提高对测控软件有缺陷模块的预测准确度,寻找支持向量机的最优参数。通过仿真实例分析测控软件有效性,并与常用缺陷预测方法进行比较,表明该模型能加快软件缺陷预测速度和提高对有缺陷模块的预测准确度。
In order to improve the prediction accuracy of software defects of support vector machine, this paper proposes a software defect prediction model based on improved support vector machine optimized by particle swarm optimization algorithm. The cost penalty coefficient is introduced to define the fitness function for PSO algorithm, and the fitness function is minimized to eliminate redundant information, to improve the software defects prediction accuracy, to find the optimal parameters of support vector machine. The validity of model is verified with data set. The simulation results show that the proposed model compared with other common defect prediction methods has improved the software defects prediction accuracy and has good nonlinear prediction ability.
出处
《计算机工程与应用》
CSCD
北大核心
2016年第11期17-21,共5页
Computer Engineering and Applications
基金
河南省科技厅发展计划(No.142102110088)
河南省科技攻关项目(No.122102210430)
关键词
缺陷预测
测控软件
粒子群优化
支持向量机
defect prediction
measure and control software
particle swarm optimization
support vector machine
