[Corpora-List] Use of NLP in CALL

[Trond Trosterud]

Hrafn Loftsson kirjoitti 29. jan. 2009 kello 14.59:

> Specifically, I'm looking for the use of PoS taggers and parsers to  
> help
> students learn, for example, word classes, morphological features,
> constituents and syntactic functions

You should check out our new program, for North Sámi, but with obvious  
possibilities for porting to other lgs:

http://oahpa.uit.no/

We use fst for the morphology, and constraint grammar for evaluating  
the student's answer in the qa and dialogue games (you may "cheat" the  
program by consulting our paradigm generator, http:// 
giellatekno.uit.no/, generate correct answers, and feed correct and  
erroneous answers, in order to evaluate. In order to evaluate the  
dialogue games you will need a Sámi speaker/learner, however).

The framework is language-independent, and we could relatively easily  
extend it to e.g. faroese, for which we have analysers (well, a bit  
more work with the analysers needed... + cooperation with pedagogues).  
As for Icelandic, IceMorph and a disambiguator is needed, and then one  
needs a component for catching the errors in the student input.

We will release our program next week, and we are not quite yet ready  
to extend it to other languages yet, but in the not too distant future  
we will.

Trond.

----------------------------------------------------------------------
Trond Trosterud                                        t +47 7764 4763
Institutt for språkvitskap, Det humanistiske fakultet  m +47 950 70140
N-9037 Universitetet i Tromsø, Noreg                   f +47 7764 5216
Trond.Trosterud (a) uit.no              http://www.hum.uit.no/a/trond/
dn------------------------------------------------------------------đŋ




_______________________________________________
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora