SVM主要是针对线性可分情况进行分析，对于线性不可分的情况，通过使用非线性映射算法将低维输入空间线性不可分的样本转化为高维特征空间使其线性可分，从而使得高维特征空间采用线性算法对样本的非线性特征进行线性分析成为可能。-SVM is mainly for the analysis of linear separable cases. Linearly indivisible cases are transformed into high dimensional feature spaces linearly by using non-linear mapping algorithm to linearly indivisible samples of low-dimensional input space, so that high-dimensional feature space It is possible to linearly analyze the nonlinear characteristics of the samples using a linear algorithm.