A numerical method that converts text into feature vectors by measuring how important each word is in a document relative to a corpus.