From key words to key semantic domains

• DOI: 10.1075/ijcl.13.4.06ray
• Publication Date: 2008-12-13T12:46:57Z
• Authors: Paul Rayson
• Keywords: Corpus Linguistics; Feature (linguistics)

This paper reports the extension of key words method for comparison corpora. Using automatic tagging software that assigns part-of-speech and semantic field (domain) tags, a is described which permits extraction domains by applying keyness calculation to tag frequency lists. The combination methods shown allow macroscopic analysis (the study characteristics whole texts or varieties language) inform microscopic level (focussing on use particular linguistic feature) thereby suggesting those features should be investigated further. resulting ‘data-driven’ approach presented here combines elements both ‘corpus-based’ ‘corpus-driven’ paradigms in corpus linguistics. A web-based tool, Wmatrix, implementing proposed applied case study: UK 2001 general election manifestos Labour Liberal Democratic parties.