Grammatical Error Generation Based on Translated Fragments

Eetu Sjöblom, Mathias Creutz, Teemu Vahtola


Abstract
We perform neural machine translation of sentence fragments in order to create large amounts of training data for English grammatical error correction. Our method aims at simulating mistakes made by second language learners, and produces a wider range of non-native style language in comparison to a state-of-the-art baseline model. We carry out quantitative and qualitative evaluation. Our method is shown to outperform the baseline on data with a high proportion of errors.
Anthology ID:
2021.nodalida-main.44
Volume:
Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa)
Month:
May 31--2 June
Year:
2021
Address:
Reykjavik, Iceland (Online)
Venue:
NoDaLiDa
SIG:
Publisher:
Linköping University Electronic Press, Sweden
Note:
Pages:
398–403
Language:
URL:
https://aclanthology.org/2021.nodalida-main.44
DOI:
Bibkey:
Cite (ACL):
Eetu Sjöblom, Mathias Creutz, and Teemu Vahtola. 2021. Grammatical Error Generation Based on Translated Fragments. In Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa), pages 398–403, Reykjavik, Iceland (Online). Linköping University Electronic Press, Sweden.
Cite (Informal):
Grammatical Error Generation Based on Translated Fragments (Sjöblom et al., NoDaLiDa 2021)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2021.nodalida-main.44.pdf
Data
OpenSubtitlesWI-LOCNESS