基于核主成分KPCA的LS-SVM的算法分析

主成分分析（PCA）是一种重要的特征抽取方法,而核主成分分析（KPCA）是在此基础上结合核函数提出来的主成分分析法,在多维回归分析中具有重要的作用。最小二乘支持向量机（LS-SVM）是SVM的一种改进算法,将KPCA与LS-SVM结合起来建模,并试验说明较之单独用最小二乘支持向量机（LS-SVM）建模方法具有更良好的推广性。

LS-SVM Regression Based on KPCA

Support vector machines(SVM)is more and more applied in the various fields in recent years.Least Square Support Vector Machines(LS-SVM) is one of evolutionary SVM.Kernel principal component analysis(KPCA) is one of most important method of feature selection and feature extract ion.This paper applies kernel principal component analysis(KPCA) to LS-SVM regression.Finally through comparing the non-KPCA method and the KPCA method mentioned in this paper separately,we could obtain that using KPCA on LS-SVM regre...