Good points. Honestly, I feel like a discussion on PCA using the SVD as a background is much more intuitive, and would fit with some of the author has claimed a

Good points. Honestly, I feel like a discussion on PCA using the SVD as a background is much more intuitive, and would fit with some of the author has claimed about eigenvectors.