Abstract
<jats:title>Abstract</jats:title> <jats:p>From a corpus perspective, keywords are those whose frequency (or infrequency) in a text or corpus is “unusual” when compared to the standards set by a reference corpus. Keywords provide insights into specific linguistic patterns, cultural meanings, and discourse practices. The calculation of “keyness” quantifies the salience of linguistic items, highlighting over‐ or under‐represented features that distinguish a text, genre, language variety, or period in the history of a language. Despite the widespread availability of tools for keyword analysis, the process remains complex, requiring careful selection of reference corpora, preprocessing of data, and interpretation of statistical measures. The entry provides the basic definitions and discusses methodological issues of corpus selection, data preprocessing, keyness calculation, and selection. Special attention is paid to the choice of the reference corpus and statistical issues, such as probability tests, effect size tests, and dispersion measures. The final part of the entry looks briefly into the applications of keywords analysis, its challenges, limitations, and future directions. The use of keyword analysis extends across various domains, including discourse analysis, stylistics, and language teaching, but important challenges regard corpus composition and keyword selection. As corpus linguistics evolves, the integration of keyness analysis with other computational methods and its application to multimodal data promise to enhance our understanding of language use.</jats:p>