The Google Korean Universal Dependency Treebank is first converted from the Universal Dependency Treebank v2.0 (legacy), and then enhanced by Chun et al., 2018.
This is a collaborative work by (in alphabetic order):
- Jinho Choi, Emory University
- Jayeol Chun, Emory University
- Na-Rae Han, University of Pittsburgh
- Jena D. Hwang, Institute for Human & Machine Cognition.
- Ryan McDonald, Google Research
- Joakim Nivre, Uppsala University
- Daniel Zeman, Institute of Formal and Applied Linguistics
The project repository: https://github.com/emorynlp/ud-korean
- Building Universal Dependency Treebanks in Korean, Jayeol Chun, Na-Rae Han, Jena D. Hwang, and Jinho D. Choi. In Proceedings of the 11th International Conference on Language Resources and Evaluation, LREC'18, Miyazaki, Japan, 2018.
- 2018-04-15 v2.2
- Significant rework, fixed some annotation errors.
- Added lemmas and fine-grained part-of-speech tags automatically generated by the KOMA morphological analyzer.
- 2017-11-15 v2.1
- No changes.
- 2017-03-01 v2.0
- Initial UD release.
=== Machine-readable metadata ================================================= Data available since: UD v2.0 License: CC BY-NC-SA 3.0 US Includes text: yes Genre: news blog Lemmas: automatic UPOS: converted from manual XPOS: automatic Features: not available Relations: converted from manual Contributors: McDonald, Ryan; Nivre, Joakim; Zeman, Daniel; Choi, Jinho; Han, Na-Rae; Hwang, Jena; Chun, Jayeol Contributing: here Contact: jinho.choi@emory.edu =============================================================================== (Original treebank contributors: LaMontagne, Adam; Souček, Milan; Järvinen, Timo; Radici, Alessandra)