Glossary math Term Page
Principal Component Analysis
A method for finding the main variance directions
Core Idea
Principal component analysis finds orthogonal directions along which data varies the most. These directions are obtained from the eigenvectors of the covariance matrix or, equivalently, through SVD on centered data.
Role In This Blog
In Mathbong, PCA is the main application that makes eigenvalues, eigenvectors, and SVD immediately useful for data analysis and dimensionality reduction.