Korpusbasert leksikonbygging

Abstract

The article outlines the motivation for two research projects relevant for computational lexicography applied to Norwegian: a text corpus and a computational dictionary. The corpus must be composed with special attention to the rather special status for written Norwegian because Norwegian has two written standards (Bokmål and Nynorsk). It is argued that these standards contain substandards which are not described systematically. We argue that such substandards can be discovered by careful use of corpora, and the substandards can be represented as different electronic dictionaries to be used in for instance proof reading systems.