LINGUIST List 14.3477

Tue Dec 16 2003

FYI: Portuguese Morpholympics Data Released

Editor for this issue: Anne Clarke <annelinguistlist.org>

Directory

1. Luis Costa, Portuguese Morpholympics Data Finally Released

Portuguese Morpholympics data released.

Linguateca is pleased to announce that data, results and programs of
the first evaluation contest for Portuguese Morfolimp�adas are now
available for download at Linguateca's site. www.linguateca.pt ->
Avalia��o conjunta -> Morfolimp�adas (and also as a tar file).

The 1st Portuguese Morpholympics's last round took place at
Avalon'2003 the 28th June 2003, at Faro, Universidade do Algarve. We
have finally managed to create a distribution with everything in
place.

The winner of the contest was Eckhard Bick's PALMORF. Seven other
systems, from Portugal and Brazil, participated in the evaluation
contest and kindly gave us the right to distribute the material.

We distribute the golden list used, the input texts, the output of
every system (after anonymization) and the programs used to compute
the results. We also provide extensive documentation (in Portuguese),
including the actual results made available already in June.

We believe that a corpus of differently tokenized and morhologically
analysed running Portuguese text is interesting for further research
in Portuguese morphology, tokenization and to improve evaluation
setups.

URLs:
http://www.linguateca.pt/Morfolimpiadas/
http://www.linguateca.pt/avalon2003/

For the organizing committee (Luis Costa, Paulo Rocha and Diana
Santos) 
Luis

Luis Costa Linguateca
 
SINTEF Telecom & Informatics
Tel. (directo) +47 22 06 73 11
Forskningsveien 1Tel. +47 22 06 73 00
Box 124 Blindern
Fax. +47 22 06 73 50
N-0314 Oslo
Email: luis.costasintef.no
Noruega
http://www.linguateca.pt/