基于PCA-ISVM的软件缺陷预测模型  

Software Defect Prediction Model Based on PCA-ISVM

作　　者：刘芳[1] 高兴[1] 周冰[1] 邓娟[1] 

机构地区：[1]武汉科技大学城市学院,湖北武汉430083

出　　处：《计算机仿真》2014年第3期397-401,共5页Computer Simulation

摘　　要：在软件缺陷检测中,常常伴有大量的冗余干扰信息,给准确检测造成困难。为了提高对软件缺陷预测准确率,提出了一种主成分分析和混沌粒子算法优化支持向量机的软件缺陷预测方法(PCA-ISVM)。首先利用主成分分析消除软件数据冗余信息,然后处理后软件数据输入到支持向量机进行训练,并通过混沌粒子群算法优化支持向量机参数,建立最优软件缺陷预测模型,最后采用仿真对模型有效性验证。仿真结果表明,提出的模型有效消除软件数据中冗余信息,获得最优支持向量机参数,从而提高了软件缺陷预测准确率和加快软件缺陷预测速度。In order to improve the prediction accuracy of software defects of support vector machine, this paper proposed a software defect prediction model based on principal component analysis and support vector machine opti- mized by chaotic particle optimization algorithm. Firstly, principal component analysis was used to eliminate redun- dant information in software defect data, and then chaotic particle swarm optimization algorithm was used to optimize the parameters of support vector machine parameter, and the support vector machine was trained to establish the opti- mal software defect prediction model. Finally, the validity of model was verified with data set. The simulation results show that the proposed model has improved the software defects prediction accuracy and has good nonlinear prediction ability.

关 键 词：软件缺陷 主成分分析 混沌粒子群优化算法 支持向量机 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]