TF-IDF

A word-count weighting that discounts words common across the whole corpus, so distinctive words stand out more than raw counts alone would show.

TF-IDF (term frequency-inverse document frequency) weights a word's raw count in one document (term frequency) by a factor that shrinks as the word appears in more documents across the whole corpus (inverse document frequency), so a word that is frequent in one document but rare corpus-wide scores higher than a word that is simply common everywhere (like "the").

Examples

The word "the" gets a near-zero TF-IDF weight because it appears in almost every document, while a rare word like "defibrillator" scores high in the few documents that mention it.