Identifying Metadata-Specific Collocations in Text Corpora

A statistical procedure based on corpus metadata to identify domain-specific collocations in an intuitive way is introduced, and a user interface connected to the word sketch feature of the Sketch Engine corpus interface is presented.