[Corpora-List] word sense disambiguation in free-order languages (fwd)

Yannick Versley versley at sfs.uni-tuebingen.de
Wed Apr 4 10:13:25 UTC 2007


Hi,

there is a LFG parser for Urdu which might be usable for syntax-based work:
http://ling.uni-konstanz.de/pages/home/butt/pargram/index.html
(you would have to ask Miriam Butt herself for possibilities of access and/or 
use of the LFG parser and grammar, and the lexical coverage may still be too 
limited for parsing unrestricted text - but increasing lexical coverage can 
be less work than building even a shallow parser yourself).
Using this, it could be possible to use Lin's syntax-based techniques for 
collocation detection and/or WSD.

On the other hand, it's perfectly possible to do WSD without using any 
syntactic information, e.g. the work of Buitelaar et al (2001) for German:
http://citeseer.ist.psu.edu/buitelaar01unsupervised.html
(This might also be an example for WSD on a language with (semi-)free word 
order, with morphological analysis being as big a headache as it probably is 
for Urdu).

Best,
Yannick
> I am doing MS(CS) and my are of specialization is NLP. I am currently
> working in area of corpus linguistics specifically issues related to Urdu.
>
> I wanted to know if there has been any work done regarding collocation and
> word sense disambiguation on Urdu in Leeds or anywhere else that is in your
> knowledge.
>
> I have tried to search for relevant material not only for Urdu but for any
> free-order language but was unable to find. Do share with me if you are
> aware of work done regarding WSD on any free order language.
-- 
Yannick Versley
Seminar für Sprachwissenschaft, Abt. Computerlinguistik
Wilhelmstr. 19, 72074 Tübingen
Tel.: (07071) 29 77352



More information about the Corpora mailing list