小波支持向量机在软件衰退预测的应用研究  

Application of Software Aging Prediction Based on Wavelet Analysis and Support Vector Machines

作　　者：朱小川[1] 

机构地区：[1]广东科学技术职业学院,广东广州510640

出　　处：《计算机仿真》2012年第3期266-269,共4页Computer Simulation

摘　　要：研究软件衰退问题,软件衰退数据存在大量噪声,传统预测方法难以消除噪声,预测精度低。为提高软件衰退预测精度,提出一种小波支持向量机的软件衰退预测方法。首先对收集软件衰退预测数据进行归一化处理,而后采用小波分析对数据进行分解,分解成多信尺度,而后采用支持向量机对软件衰退数据各个尺度系数分别进行预测,最后采用小波分析对各尺度系数预测结果进行重构,得到软件衰退预测的最终结果。仿真结果表明,相对传统预测方法,小波支持向量机提高了软件衰退预测精度,能够很好地预测软件衰退趋势。Software aging data contain a lot of noises,and the traditional forecasting method can not effectively eliminate the noises.In order to improve the prediction accuracy of software aging,the paper presented a software aging prediction method based on wavelet support vector machine.The collection of software aging prediction data was reduced,then wavelet analysis was performed on the data decomposition which is of multiple letter scales.Then,software aging data of various scales coefficients were predicted by using wavelet analysis.Last,the obtained coefficient forecasting results were used for reconstruction to get the software aging prediction result.The simulation results show that,compared with the traditional forecasting methods,wavelet support vector machine can improve the software aging prediction accuracy and predict the trend of decline of software.

关 键 词：软件衰退 小波分析 支持向量机 预测 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]