UIT The arctic university of Norway > Giellatekno

What Is This

This is a morphological analyser for Faroese. It is based upon, and contains the lemmas from Føroysk orðabók. The analyser is still a bit rough-edged, but it is good enough to be useful.

The source code is available as open source (see the links on the previous page).

The parser is a finite state transducer, capable of analysing and generating any Faroese wordform. The parser is still under development. In november 2006, it recognised 85 % of the wordforms in a corpus containing appr. 2.35 million words. The grammatical part of the parser is written at the University of Tromsø, and the lexicon is the set of lexemes found in the two-volume Føroysk orðabók (Føroya Fróðskaparfelag 1998).

The project also contains a morphological disambiguator, built upon constraint grammar technology. The Faroese disambiguator is still in an embryonic state (as you can verify by testing it out).

For a presentation, see Trond Trosterud: A Constraint Grammar for Faroese.

The structure of the source file catalogue is the same for all the languages, it is documented here.

These pages document how to build the parsers. If you want to know how to use them on the command line, read the How to use the morphological parsers page.