tf–idf

In <a href="/facts/Information_retrieval/MFSmpeNw">information retrieval</a>, tf–idf (also TF*IDF, TFIDF, TF–IDF, or Tf–idf), short for term frequency–inverse document frequency, is a measure of importance of a word to a <a href="/facts/Document/H4Hdxu91">document</a> in a collection or <a href="/facts/Text_corpus/bYeEQQdh">corpus</a>, adjusted for the fact that some words appear more frequently in general. Like the bag-of-words model, it models a document as a <a href="/facts/Multiset/bPFJGMDo">multiset</a> of words, without <a href="/facts/Word_order/nXtB9Gm2">word order</a>. It is a refinement over the simple <a href="/facts/Bag-of-words_model/TlhIx600">bag-of-words model</a>, by allowing the weight of words to depend on the rest of the corpus.
It was often used as a <a href="/facts/Weighting_factor/HUltmeDW">weighting factor</a> in searches of information retrieval, <a href="/facts/Text_mining/5QsIa3Cz">text mining</a>, and <a href="/facts/User_modeling/uVXnQ9S6">user modeling</a>. A survey conducted in 2015 showed that 83% of text-based recommender systems in digital libraries used tf–idf. Variations of the tf–idf weighting scheme were often used by <a href="/facts/Search_engine/SRntsbz2">search engines</a> as a central tool in scoring and ranking a document's <a href="/facts/Relevance_(information_retrieval)/0kkQ6B9A">relevance</a> given a user <a href="/facts/Information_retrieval/MFSmpeNw">query</a>.
One of the simplest <a href="/facts/Ranking_function/VC9wFqwx">ranking functions</a> is computed by summing the tf–idf for each query term; many more sophisticated ranking functions are variants of this simple model.

tf–idf open-in-new

tf–idf