在多元统计领域中，核函数主成分分析(kernel principal component analysis, kernel PCA)是利用核函数方法技术对主成分分析(PCA)的扩展。使用核函数使原PCA的线性操作是在一个复制的内核希尔伯特空间中执行的。 KPCA的运算步骤势在PCA之前首先对数据进行kernel变换 ，再求相关系数矩阵。(In the field of multivariate statistics, kernel principal component analysis (kernel PCA) is an extension of principal component analysis (PCA) using kernel function technology. The linear operation of the original PCA is performed in a replicated kernel Hilbert space using a kernel function. The operation step potential of KPCA is to transform the data into kernels before PCA, and then calculate the correlation coefficient matrix.)