Bringing data and dictionary together: Real science in real dictionaries

This chapter looks at how general language dictionaries may be extended by using prototype theory to take into account specialised usage, and how a specialised pattern dictionary can be built bottom-up from a corpus. Following an analysis of the role of science in current learner's dictionaries, a study of how existing entries can be improved using corpus data and lexicographical prototypes will be shown. The main thrust is a corpus-driven approach to extending current dictionaries so as to highlight the particular production environment of the scientific article and also to build a new dictionary using collocational networks for both word selection and as an exploratory tool. Such a dictionary is termed 'organic' in that it grows naturally from the corpus data