BERT-based Automatic Error Correction System for Chinese Learners
Main Article Content
Abstract
Current automatic error correction methods for Chinese learners often focus on superficial word- or sentence-level processing and are affected by inconsistent annotation standards, resulting in limited generalization to real learner texts and causing misalignment or overcorrection. To improve grammatical correctness, semantic fidelity, and instructional relevance, this paper constructs a multi-granularity lexical representation method. Using characters as the basic unit, the method integrates three types of information: characters, words, and pinyin. These representations are learned through projection processing, concatenated into a unified embedding, and fed into a pre-trained Chinese BERT encoder. A dual-task head consisting of sequence labeling and lightweight generation is deployed on the shared BERT encoder for collaborative optimization. Key innovations include an alignment consistency loss to ensure character-level consistency, and the combination of word-segmentation augmentation and multi-reference soft-label training to reduce conflicts caused by segmentation differences. Experimental results show that the method achieves granular alignment consistency above 0.800, corrected-original sentence similarity of 0.888, and an overcorrection rate as low as 0.041. These results demonstrate improved robustness and explainability for automatic Chinese learner error correction.
Downloads
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
References
Wang, H, Z. Chen, Z. Zhang, Z. Ling, X. Pan, W. Duan, et al., “Revisiting the Evaluation for Chinese Grammatical Error Correction,” Journal of Advanced Computational Intelligence and Intelligent Informatics, vol. 28, no. 6, pp. 1380-1390, 2024, doi: 10.20965/jaciii.2024.p1380.
Q. Ao, “The sociocultural-cognitive underpinnings in error correction: A descriptive study on acquisition of Chinese tones,” International Journal of Chinese Language Teaching, vol. 2, no. 2, pp. 44-72, 2021, doi: 10.46451/ijclt.2021.10.04.
R. Shadiev and Y. Feng, “Using automated corrective feedback tools in language learning: A review study,” Interactive Learning Environments, vol. 32, no. 6, pp. 2538-2566, 2024, doi: 10.1080/10494820.2022.2153145.
F. Zhu, “Supporting EFL writing during the pandemic: The effectiveness of data-driven learning in error correction,” Asian EFL Journal, vol. 25, no. 5, pp. 8-27, 2021.
C. Bryant, Z. Yuan, M. R. Qorib, H. Cao, H. T. Ng, and T. Briscoe, “Grammatical error correction: A survey of the state of the art,” Computational Linguistics, vol. 49, no. 3, pp. 643-701, 2023, doi: 10.1162/coli_a_00478.
Z. Sun, K. Yu, and X. Wu, “A Chinese Error Correction Algorithm Based on the Fusion of Text Sequence Error Probability and Chinese Spelling Error Probability,” Application Research of Computers, vol. 40, no. 8, pp. 2292-2297, 2023.
W. Gou and Z. Chen, “Think twice: A post-processing approach for the Chinese spelling error correction,” Applied Sciences, vol. 11, no. 13, pp. 5832-5847, 2021, doi: 10.3390/app11135832.
Y. Wang, Y. Wang, K. Dang, J. Liu, and Z. Liu, “A comprehensive survey of grammatical error correction,” ACM Transactions on Intelligent Systems and Technology (TIST), vol. 12, no. 5, pp. 1-51, 2021, doi: 10.1145/3474840.
M. Qin, “A study on automatic correction of English grammar errors based on deep learning,” Journal of Intelligent Systems, vol. 31, no. 1, pp. 672-680, 2022, doi: 10.1515/jisys-2022-0052.
Y. Zhong and X. Yue, “On the correction of errors in English grammar by deep learning,” Journal of Intelligent Systems, vol. 31, no. 1, pp. 260-270, 2022, doi: 10.1515/jisys-2022-0013.
S. Zhu, J. Liu, Y. Li, and Z. Yu, “Automatical sampling with heterogeneous corpora for grammatical error correction,” Complex & Intelligent Systems, vol. 11, no. 1, pp. 25-35, 2025, doi: 10.1007/s40747-024-01653-3.
R. Duan, Z. Ma, Y. Zhang, Z. Ding, and X. Liu, “Dynamic Assessment-Based Curriculum Learning Method for Chinese Grammatical Error Correction,” Electronics, vol. 13, no. 20, pp. 4079-4091, 2024, doi: 10.3390/electronics13204079.
A. Musyafa, Y. Gao, A. Solyman, C. Wu, and S. Khan, “Automatic correction of indonesian grammatical errors based on transformer,” Applied Sciences, vol. 12, no. 20, pp. 10380-10396, 2022, doi: 10.3390/app122010380.
N. M. Gardazi, A. Daud, M. K. Malik, A. Bukhari, T. Alsahfi, and B. Alshemaimri, “BERT applications in natural language processing: A review,” Artificial Intelligence Review, vol. 58, no. 6, pp. 1-49, 2025, doi: 10.1007/s10462-025-11162-5.
Y. Cho, “Grammatical illusions in BERT: Attraction effects of subject-verb agreement and reflexive-antecedent dependencies,” Linguistic Research, vol. 40, no. 2, pp. 317-352, 2023, doi: 10.17250/khisli.40.2.202306.007.
I. Papadimitriou, E. A. Chi, R. Futrell, and K. Mahowald, “Multilingual BERT, ergativity, and grammatical subjecthood,” Society for Computation in Linguistics, vol. 4, no. 1, pp. 425-426, 2021, doi: 10.7275/6p1w-g122.
Z. Wang, Q. Yu, J. Wang, Z. Hu, and A. Wang, “Grammar Correction for Multiple Errors in Chinese Based on Prompt Templates,” Applied Sciences, vol. 13, no. 15, pp. 8858-8874, 2023, doi: 10.3390/app13158858.
H. Wang, M. Kurosawa, S. Katsumata, M. Mita, and M. Komachi, “Chinese grammatical error correction using pretrained models and pseudo data,” ACM Transactions on Asian and Low-Resource Language Information Processing, vol. 22, no. 3, pp. 1-12, 2023, doi: 10.1145/3570209.
Y. Qing, “Design and application of automatic english translation grammar error detection system based on bert machine vision,” Scalable Computing: Practice and Experience, vol. 25, no. 3, pp. 2088-2102, 2024, doi: 10.12694/scpe.v25i3.2770.
C. Mi, S. Xie, and Y. Fan, “Multi-granularity Knowledge Sharing in Low-resource Neural Machine Translation,” ACM Transactions on Asian and Low-Resource Language Information Processing, vol. 23, no. 2, pp. 1-19, 2024, doi: 10.1145/3639930.
M. Xin, L. Ma, and B. Hu, “Research on text classification by integrating multi-granularity information,” Journal of Computer Engineering & Applications, vol. 59, no. 9, pp. 104-111, 2023.
W. Qiu and J. Xu, “A multi-granularity relationship detection model integrating global and local features,” Application Research of Computers, vol. 40, no. 2, pp. 476-480, 2023.
W. Jin, F. Jiang, X. Wang, N. Ma, and Y. Zhang, “Research and analysis of grammatical error correction technology for Chinese documents,” Journal of Computer and Communications, vol. 12, no. 8, pp. 202-223, 2024.
H. Xie, Z. Chen, J. Cheng, X. Lü, and Z. Tang, “Chinese Grammatical Error Detection Method Based on Data Augmentation and Multi-task Feature Learning,” Journal of Chinese Information Processing, vol. 36, no. 12, pp. 36-43, 2022, [Online]. Available: https://aclanthology.org/2020.ccl-1.71/.
R. Yang, Y. Gan, and C. Zhang, “Chinese named entity recognition based on BERT and lightweight feature extraction model,” Information, vol. 13, no. 11, pp. 515-535, 2022, doi: 10.3390/info13110515.
B. Qiao, Z. Zou, Y. Huang, K. Fang, X. Zhu, and Y. Chen, “A joint model for entity and relation extraction based on BERT,” Neural Computing and Applications, vol. 34, no. 5, pp. 3471-3481, 2022, doi: 10.1007/s00521-021-05815-z.
G. S. Chauhan, Y. K. Meena, D. Gopalani, and R. Nahta, “A mixed unsupervised method for aspect extraction using BERT,” Multimedia Tools and Applications, vol. 81, no. 22, pp. 31881-31906, 2022, doi: 10.1007/s11042-022-13023-7.
K. H. Alyoubi, F. S. Alotaibi, A. Kumar, V. Gupta, and A. Sharma, “A novel multi-layer feature fusion-based BERT-CNN for sentence representation learning and classification,” Robotic Intelligence and Automation, vol. 43, no. 6, pp. 704-715, 2023, doi: 10.1108/RIA-04-2023-0047.
J. Su, S. Yu, and X. Hong, “A self-supervised pre-training method for Chinese spelling correction,” Journal of South China University of Technology (Natural Science Edition), vol. 51, no. 9, pp. 90-98, 2023, doi: 10.12141/j.issn.1000-565X.230031.
S. Kumar and A. Solanki, “An abstractive text summarization technique using transformer model with self-attention mechanism,” Neural Computing and Applications, vol. 35, no. 25, pp. 18603-18622, 2023, doi: 10.1007/s00521-023-08687-7.
Y. Zhang, H. Kamigaito, and M. Okumura, “Bidirectional transformer reranker for grammatical error correction,” Journal of Natural Language Processing, vol. 31, no. 1, pp. 3-46, 2024, doi: 10.5715/jnlp.31.3.
C. Liu, K. Zhang, M. Bao, Y. Liu, and Q. Liu, “A Chinese Spelling Correction Method Based on the Fusion of Context and Text Structure,” Journal of Nanjing University (Natural Science Edition), vol. 60, no. 3, pp. 451-463, 2024.
G. Huang, C. Wang, and Y. Zhang, “A Chinese Text Correction Method Based on Dynamic Text Window and Dynamic Weight Allocation,” Journal of Zhengzhou University (Science Edition), vol. 52, no. 3, pp. 9-14, 2020.
Y. Zha, Y. Yang, R. Li, and Z. Hu, “Text alignment is an efficient unified model for massive nlp tasks,” Advances in Neural Information Processing Systems, vol. 36, pp. 77942-77968, 2023.
X. Xie, G. Xu, L. Zhao, and R. Guo, “Opensearch-sql: Enhancing text-to-sql with dynamic few-shot and consistency alignment,” Proceedings of the ACM on Management of Data, vol. 3, no. 3, pp. 1-24, 2025, doi: 10.1145/3725331.