Term extraction is the automated identification of candidate terms in a document, based on how frequently word sequences occur. The candidates are reviewed and stored in a glossary for later use in translation and terminology work.