Training a Swedish Constituency Parser on Six Incompatible Treebanks

Richard Johansson, Yvonne Adesam


Abstract
We investigate a transition-based parser that uses Eukalyptus, a function-tagged constituent treebank for Swedish which includes discontinuous constituents. In addition, we show that the accuracy of this parser can be improved by using a multitask learning architecture that makes it possible to train the parser on additional treebanks that use other annotation models.
Anthology ID:
2020.lrec-1.642
Volume:
Proceedings of the Twelfth Language Resources and Evaluation Conference
Month:
May
Year:
2020
Address:
Marseille, France
Venue:
LREC
SIG:
Publisher:
European Language Resources Association
Note:
Pages:
5219–5224
Language:
English
URL:
https://aclanthology.org/2020.lrec-1.642
DOI:
Bibkey:
Cite (ACL):
Richard Johansson and Yvonne Adesam. 2020. Training a Swedish Constituency Parser on Six Incompatible Treebanks. In Proceedings of the Twelfth Language Resources and Evaluation Conference, pages 5219–5224, Marseille, France. European Language Resources Association.
Cite (Informal):
Training a Swedish Constituency Parser on Six Incompatible Treebanks (Johansson & Adesam, LREC 2020)
Copy Citation:
PDF:
https://preview.aclanthology.org/remove-xml-comments/2020.lrec-1.642.pdf