FYI: 2nd International Chinese Word Segmentation Bakeoff
The Second International Chinese Word Segmentation Bakeoff
Preliminary Description and Important Dates
This is the initial announcement for the Second International Chinese Word
Segmentation Bakeoff, sponsored by the Special Interest Group for Chinese
Language Processing (SIGHAN) of the Association for Computational
Linguistics. The bakeoff will occur over the summer of 2005 and the results
will be presented at the 4th SIGHAN Workshop, to be held at IJCNLP'05,
The first bakeoff, held in 2003 and presented at the 2nd SIGHAN Workshop at
ACL 2003 in Sapporo, has become the pre-eminent measure for Chinese word
segmentation evaluation and has been cited in numerous papers. As with the
first evaluation, the second bakeoff will concentrate exclusively on Word
Segmentation. Corpora from the following organizations will be available
- CKIP, Academia Sinica, Taiwan
- City University of Hong Kong, Hong Kong SAR
- CIS Department, University of Pennsylvania, United States
- Beijing Universty, China
- Microsoft Research, China
The exact nature of the segmentation tasks is being discussed and final
details will be made available when registration is opened on 1 June 2005.
Participants are required to submit a short paper describing their system
and analyzing their performance, and present a summary at the workshop. The
reports will be published in the SIGHAN workshop proceedings.
The language of the workshop is English. Papers must be submitted and
presented in English. Note that unlike the workshop proper, there will not
be a peer review process on the bakeoff reports.
2. Important Dates
2005-06-01 Registration Open
2005-06-29 Training data made available
2005-07-27 Testing data made available
2005-07-29 Test results sent back to organizers
2005-08-05 Results privately reported to participants
2005-08-19 Final reports due from participants
3. Contact Information
The workshop is being organized by Tom EMERSON of Basis Technology Corp.
and Jianfeng GAO of Microsoft Research China.
The web page for the competition is:
Questions on the bakeoff should be addressed to Tom Emerson,