PCA
Also: principal component analysis
Finding the directions along which data varies most, and describing each point by its position along the first few. Compresses many correlated columns into a handful of uncorrelated ones.
A definition is not an understanding. This is taught properly in PCA from scratch, part of Linear algebra.
Learn this properlyTerms from the same topic