Tf-idf

Tf-idf, short for term frequency-inverse document frequency, is a measure used in information retrieval of how important a word is to a document within a collection or corpus, adjusted for the fact that some words appear more often across documents in general. Like bag of words models it treats a document as a multiset of words without regard to order, but improves on that approach by letting word weights vary according to the whole corpus rather than treating every word the same. It is widely used in information retrieval, text mining and user modeling; a 2015 survey found that roughly 83 percent of text based recommender systems in digital libraries used it, and search engines commonly rely on variations of the scheme to score and rank documents against a query, often by summing the tf-idf scores of the query's own terms. This description is adapted from Wikipedia contributors under CC BY-SA 4.0; changes were made. https://creativecommons.org/licenses/by-sa/4.0/