This document discusses matrix factorization techniques. Matrix factorization involves decomposing a large, sparse matrix into more compact, ordered matrices. Common factorization methods include singular value decomposition (SVD), principal component analysis (PCA), and non-negative matrix factorization (NMF). NMF decomposes a matrix into two lower rank, non-negative matrices. It can reduce space needed to store data compared to SVD. The document provides an example of using NMF to factorize a user-item rating matrix to make recommendations.