Info and questions about text corpora

Santos Diana | 1 Mar 2004 13:48

RE: NLP tools for Portuguese?

Dear João,

At www.linguateca.pt we maintain a catalogue of NLP tools for Portuguese
(Catálogo de ferramentas) as well as a developer's oriented site to actually
get the publicly available ones.

We also maintain a publications catalogue about NLP dealing with Portuguese
(Catálogo de publicações), currently with 742 entries.

One of our main lines of activity is evaluation of Portuguese NLP, and we
maintain a list devoted to this subject (avalia <at> linguateca.pt) to which I
suggest you subscribe, as well as a reference list on evaluation of
Portuguese tools and systems (see Avaliação conjunta).

I use this opportunity to ask members of this list who have developed
systems that have been applied to Portuguese, and/or have published on this
subject, and who are not yet mentioned in our site to help us enriching the
catalogues above.

Saudações,
Diana
============
Diana Santos
www.linguateca.pt
Linguateca, Oslo node, SINTEF ICT
Pb 124 Blindern, N-0314 Oslo, Norway

-----Original Message-----
From: jsilva <at> esoterica.pt [mailto:jsilva <at> esoterica.pt] 
Sent: 27. februar 2004 14:49

(Continue reading)

headers

Tony Berber Sardinha | 29 Feb 2004 15:13

NLP tools for Portuguese?

Dear João

Please have a look at http://lael.pucsp.br/corpora where I've put up several
tools for tagging, chunking, concordancing, aligning, keywording and more

cheers
tony.
-------------------------------------
Dr Tony Berber Sardinha
LAEL, PUC/SP
(Catholic University of Sao Paulo, Brazil)
tony4 <at> uol.com.br
http://lael.pucsp.br/~tony
[New website]

> ----- Original Message -----
> From: <jsilva <at> esoterica.pt>
> To: <corpora <at> hd.uib.no>
> Sent: sexta-feira, 27 de fevereiro de 2004 10:48
> Subject: [Corpora-List] NLP tools for Portuguese?
>
>
> > Hello,
> >
> >  I am compiling an overview of existing tools for tagging
> > Portuguese. I would be grateful if you could help me by
> > letting me know about sentence chunkers, tokenizers and/or
> > pos-taggers. I am interested in knowing about the size of
> > the training data (if any), evaluation figures, evaluation
> > methods used, type of algorithm/approach (if applicable),

(Continue reading)