[Corpora-List] Help

[Diana Maynard]

It's a few years old now, but it's probably worth having a look at the 
TIDES Surprise Language Exercise.  In particular, the dry run was on 
Cebuano, a language for which few resources were available (at least at 
the time).

For a general description of the exercise, see
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.151.3647&rep=rep1&type=pdf

Google "TIDES Surprise Language Exercise" and you'll get a bunch of 
relevant papers about work done as part of the exercise, e.g.
  "NE recognition without training data on a language you don't speak" 
http://gate.ac.uk/sale/acl03/surprise.pdf

Regards
Diana


On 27/06/11 13:51, Inès Zribi wrote:
> Dear Francis
>
> Thank you for your response.
>
> I'm not interested in a particular group of languages. I am looking for
> studying methods that deal with languages that are characterized by the
> lack or even the absence of resources.
>
> Best wishes.
>
> Inès.
>
> 2011/6/27 Francis Tyers <ftyers at prompsit.com <mailto:ftyers at prompsit.com>>
>
>     El dl 27 de 06 de 2011 a les 11:05 +0100, en/na Inès Zribi va escriure:
>      > Dear Corpora list,
>      >
>      > Does anyone know any works (morphological analysis, parsing,
>      > tokenization, POS tagging, etc.) deal with under-resourced languages?
>
>     Any language or group of languages in particular ?
>
>     Fran


_______________________________________________
UNSUBSCRIBE from this page: http://mailman.uib.no/options/corpora
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora