# Principal component analysis

Principal component analysis (PCA) 叫主成分分析。它在维基百科里的定义是:

Principal component analysis is the process of computing the principal components and using them to perform a change of basis on the data, sometimes using only the first few principal components and ignoring the rest.

这里提到两个过程: 1. 计算出主成分; 2. 仅保留前几个主成分。

马上提出问题,什么是主成分? 如何找主成分?

找主成分的方法叫做 Singular Value Decomposition (SVD)

  1. Principal component analysis
  2. StatQuest: Principal Component Analysis (PCA), Step-by-Step