The Swedish Academy Dictionary, SAOB, has started a project with the aim to digitize all volumes of SAOB again to improve the quality of the text over the previous effort. The new version will first of all be correct as regards the actual characters in the text, also the typography will be identified correctly in most cases. In the second part of the project we attempt to automatically identify structures in the articles such as title, part of speech, etymologi, division of definitions and so on.
