Abstract
We introduce unsupervised techniques based on phrase-based statistical machine transla- tion for grammatical error correction (GEC) trained on a pseudo learner corpus created by Google Translation. We verified our GEC sys- tem through experiments on a low resource track of the shared task at Building Educa- tional Applications 2019 (BEA2019). As a re- sult, we achieved an F0:5 score of 28.31 points with the test data.
Cite
CITATION STYLE
Katsumata, S., & Komachi, M. (2019). (almost) unsupervised grammatical error correction using a synthetic comparable corpus. In ACL 2019 - Innovative Use of NLP for Building Educational Applications, BEA 2019 - Proceedings of the 14th Workshop (pp. 134–138). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/w19-4413
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
