Abstract
Selecting appropriate words to compose a sentence is one common problem faced by non-native Chinese learners. In this paper, we propose (bidirectional) LSTM sequence labeling models and explore various features to detect word usage errors in Chinese sentences. By combining CWINDOW word embedding features and POS information, the best bidirectional LSTM model achieves accuracy 0.5138 and MRR 0.6789 on the HSK dataset. For 80.79% of the test data, the model ranks the ground-truth within the top two at position level.- Anthology ID:
- P17-2064
- Volume:
- Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
- Month:
- July
- Year:
- 2017
- Address:
- Vancouver, Canada
- Editors:
- Regina Barzilay, Min-Yen Kan
- Venue:
- ACL
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 404–410
- Language:
- URL:
- https://aclanthology.org/P17-2064
- DOI:
- 10.18653/v1/P17-2064
- Cite (ACL):
- Yow-Ting Shiue, Hen-Hsen Huang, and Hsin-Hsi Chen. 2017. Detection of Chinese Word Usage Errors for Non-Native Chinese Learners with Bidirectional LSTM. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 404–410, Vancouver, Canada. Association for Computational Linguistics.
- Cite (Informal):
- Detection of Chinese Word Usage Errors for Non-Native Chinese Learners with Bidirectional LSTM (Shiue et al., ACL 2017)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-4/P17-2064.pdf