We review the two essentials of principal component analysis (“PCA”): 1) The principal components of a set of data points are the eigenvectors of the correlation matrix of these points in feature space. 2) Projecting the data onto the subspace spanned by the first $k$ of these — listed in descending eigenvalue order — provides the best possible $k$-dimensional approximation to the data, in the sen
![Principal component analysis – EFavDB](https://cdn-ak-scissors.b.st-hatena.com/image/square/988aeedd60cfea57ed0fda1a8322904df9f3a96d/height=288;version=1;width=512/http%3A%2F%2Fefavdb.com%2Fwp-content%2Fuploads%2F2015%2F12%2Fprojection.png)