This project aims to develop a low cost, (semi)automatic method for preparation of bi- and multilingual dictionaries. The method based on natural language processing tools and parallel corpora. The output of the pilot project is a lexicographical tool which can be a substantial aid in dictionary writing: the proto-dictionary generated automatically based on real language data contains all lexicographical information needed for lexicographers' work.

Demonstration site of EFNILEX proto-dictionaries created for Dutch-French, Hungarian-English, Hungarian-Slovenian and Hungarian-Lithuanian language pairs.