Corpus Query Tools for lexicography

Programs that extract information from corpora and display it in different ways are fundamental tools in contemporary lexicography. This article surveys the fundamental features that a state-of-the-art corpus query program should offer to lexicographers at different stages of their work, including options and issues related to tool set-up, extracting frequency information, finding collocations and inspecting keywords in context. We focus on what we think are the most important functionalities and desiderata for current and future tools, while also providing short descriptions of four popular and representative programs (TextSTAT, AntConc, the Sketch Engine, and the Corpus WorkBench). We conclude with some general considerations about user-friendliness, usability and the increasing role that NLP-based automation plays in corpus query tools.