FrameNet Annotations Alignment using Attention-based Machine Translation

Gabriel Marzinotto


Abstract
This paper presents an approach to project FrameNet annotations into other languages using attention-based neural machine translation (NMT) models. The idea is to use an NMT encoder-decoder attention matrix to propose a word-to-word correspondence between the source and the target language. We combine this word alignment along with a set of simple rules to securely project the FrameNet annotations into the target language. We successfully implemented, evaluated and analyzed this technique on the English-to-French configuration. First, we analyze the obtained FrameNet lexicon qualitatively. Then, we use existing French FrameNet corpora to assert the quality of the translation. Finally, we trained a BERT-based FrameNet parser using the projected annotations and compared it to a BERT baseline. Results show substantial improvements in the French language, giving evidence to support that our approach could help to propagate FrameNet data-set on other languages.
Anthology ID:
2020.framenet-1.6
Volume:
Proceedings of the International FrameNet Workshop 2020: Towards a Global, Multilingual FrameNet
Month:
May
Year:
2020
Address:
Marseille, France
Editors:
Tiago T. Torrent, Collin F. Baker, Oliver Czulo, Kyoko Ohara, Miriam R. L. Petruck
Venue:
Framenet
SIG:
Publisher:
European Language Resources Association
Note:
Pages:
41–47
Language:
English
URL:
https://aclanthology.org/2020.framenet-1.6
DOI:
Bibkey:
Cite (ACL):
Gabriel Marzinotto. 2020. FrameNet Annotations Alignment using Attention-based Machine Translation. In Proceedings of the International FrameNet Workshop 2020: Towards a Global, Multilingual FrameNet, pages 41–47, Marseille, France. European Language Resources Association.
Cite (Informal):
FrameNet Annotations Alignment using Attention-based Machine Translation (Marzinotto, Framenet 2020)
Copy Citation:
PDF:
https://preview.aclanthology.org/landing_page/2020.framenet-1.6.pdf