Project description
|
STO is the most comprehensive computational lexicon of Danish, it is developed
for NLP/HLT applications. The descriptive model and language is closely related
to the PAROLE/SIMPLE 'de facto standards', but he model has been
revised and the information content is enriched with very precise
language-specific information.
Size: 81,500 lemmas; Information encoded: morphology: all entries, syntax:
45,000 entries, semantics: 10,000 entries, resp.
The STO project was a national collaborational project between four expert
institutes* within the field, and it was funded by the Danish Ministry for
Science, Technology and Development (MVTU).
An internet-based user interface on the homepage of CST provides access for
searches in the STO database (the interface language is p.t. Danish only.) The
development of this interface was funded by the Electronic Research Library of
Denmark (DEFF).
Project partners were:
* Univ. of Copenhagen, Center for Sprogteknologi (CST; project management)
* Copenhagen Business School, Inst. for Computational Linguistics,
* Univ. of Copenhagen, Inst. for General and Applied Linguistics,
* Univ. of Southern Denmark, Inst. of Business Communication and
Information
Science.
The lexicon is now ready for applications. Fur further information please visit
our homepage (URL). |