Elamur Research⌕
☰
home / papers / 2018 / A Hybrid Approach to Automatic Corpus Generation for Chinese Spelling Check

A Hybrid Approach to Automatic Corpus Generation for Chinese Spelling Check

Paper · 2018 · Dingmin Wang, Yan Song, Jing Li et al.

Summary

Chinese spelling check (CSC) has lagged behind other NLP tasks because large annotated corpora are scarce, so most systems rely on unsupervised language-model methods. The authors build training data automatically by generating realistic errors with recognition tools. For visually similar (V-style) errors, they render characters as images, blur part of each image, and keep the substitutions that the OCR engine produces. A stroke-based edit distance filter then removes substitutions that look nothing like the original. For phonologically similar (P-style) errors, they convert sentences to speech and run ASR, keeping only clean substitution cases. Human evaluation found that these errors resemble mistakes people actually make. They frame CSC as sequence labeling and train a BiLSTM tagger on the generated corpus. On three standard test sets, it outperformed the same model trained on the benchmarks' own training data. They also derive confusion sets from the generated errors. The work shows that cheap synthetic error data can make supervised CSC practical.

Key points

  • OCR pipeline (D-ocr):
    • Uses Google Tesseract on 100×100 images in which a region is partly Gaussian-blurred.
    • One or two target characters are chosen per sentence.
    • Source text is 50,000 People's Daily sentences.
    • The result is about 40,000 annotated sentences containing 56,857 errors.
  • Shape filter: a substitution is rejected if the stroke-level edit distance exceeds 0.25 × the combined stroke counts of the two characters.
  • ASR pipeline: text-to-speech is followed by speech recognition. Outputs are discarded if they change sentence length, are not P-style errors, or contain too many errors.
  • Confusion sets: tested with a tri-gram LM for correction.
    • Their F1 is higher than Con1 and slightly lower than Con2.
    • They are much smaller than Con2 and were the fastest in every run.
  • Error analysis and release:
    • Most false positives occur in idioms and poems.
    • Errors whose characters are neither visually nor phonetically similar are missed (e.g., 禁/戒).
    • Code and data are published at github.com/wdimmy/Automatic-Corpus-Generation.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🏢TencentShenzhen · China66 works · 17,945 citations

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 9,086 worksArtificial Intelligence · 8,889 worksArts and Humanities · 9,296 worksLanguage and Linguistics · 9,206 worksComputer Vision and Pattern Recognition · 1,282 works

Cite

Cite as Dingmin Wang, Yan Song, Jing Li, Jialong Han and Haisong Zhang. 2018. A Hybrid Approach to Automatic Corpus Generation for Chinese Spelling Check. Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/d18-1273

@inproceedings{wang2018hybrid,
  title = {{A Hybrid Approach to Automatic Corpus Generation for Chinese Spelling Check}},
  author = {Dingmin Wang, Yan Song, Jing Li, Jialong Han and Haisong Zhang},
  booktitle = {Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing},
  year = {2018},
  doi = {10.18653/v1/d18-1273},
  url = {https://doi.org/10.18653/v1/d18-1273}
}

References

46 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Gary Bradski. 2000. The OpenCV Library. Dr. Dobb’s Journal: Software Tools for the Professional Programmer, 25(11):120–123. find
  2. Keith P Braho, Jeffrey P Pike, and Lori A Pike. 2014. Methods And Systems for Identifying Errors in A Speech Recognition System. US Patent 8,868,421. find
  3. Hui Bu, Jiayu Du, Xingyu Na, Bengu Wu, and Hao Zheng. 2017. AIShell-1: An Open-source Mandarin Speech Corpus and A Speech Recognition Baseline. arXiv preprint arXiv:1709.05522. — [arXiv:1709.05522](https://arxiv.org/abs/1709.05522) find
  4. Jill Burstein and Martin Chodorow. 1999. Automated Essay Scoring for Nonnative English Speakers. In Proceedings of a Symposium on Computer Mediated Language Assessment and Evaluation in Natural Language Processing, pages 68–75. Association for Computational Linguistics. find
  5. Chao-Huang Chang. 1995. A New Approach for Automatic Chinese Spelling Correction. In Proceedings of Natural Language Processing Pacific Rim Symposium, volume 95, pages 278–283. find
  6. Chao-Huang Chang et al. 1993. HMM-based Part-of-speech Tagging for Chinese Corpora. Very Large Corpora: Academic and Industrial Perspective. find
  7. Tao-Hsing Chang, Hsueh-Chih Chen, Yuen-Hsien Tseng, and Jian-Liang Zheng. 2013. Automatic Detection and Correction for Chinese Misspelled Words Using Phonological and Orthographic Similarities. In Proceedings of the Seventh SIGHAN Workshop on Chinese Language Processing, pages 97–101. find
  8. Kuan-Yu Chen, Hung-Shin Lee, Chung-Han Lee, Hsin-Min Wang, and Hsin-Hsi Chen. 2013. A study of Language Modeling for Chinese Spelling Check. In Proceedings of the 7th SIGHAN Workshop on Chinese Language Processing (SIGHAN-13), pages 79– 83. find
  9. Yong-Zhi Chen, Shih-Hung Wu, Ping-Che Yang, Tsun Ku, and Gwo-Dong Chen. 2011. Improve the Detection of Improperly Used Chinese Characters in Students’ Essays with Error Model. International Journal of Continuing Engineering Education and Life Long Learning, 21(1):103–116. find
  10. Shichao Dong, Gabriel Pui Cheong Fung, Binyang Li, Baolin Peng, Ming Liao, Jia Zhu, and Kam-Fai Wong. 2016. ACE: Automatic Colloquialism, Typographical and Orthographic Errors Detection for Chinese Language. In COLING (Demos), pages 194–197. find
  11. Gabriel Fung, Maxime Debosschere, Dingmin Wang, Bo Li, Jia Zhu, and Kam-Fai Wong. 2017. find
  12. NLPTEA 2017 Shared Task–Chinese Spelling Check. In Proceedings of the 4th Workshop on Natural Language Processing Techniques for Educational Applications (NLPTEA 2017), pages 29–34. find
  13. Jianfeng Gao, Xiaolong Li, Daniel Micol, Chris Quirk, and Xu Sun. 2010. A Large Scale Ranker-based System for Search Query Spelling Correction. In Proceedings of the 23rd International Conference on Computational Linguistics, pages 358–366. Association for Computational Linguistics. find
  14. Alex Graves. 2013. Generating Sequences With Recurrent Neural Networks. arXiv pre-print, abs/1308.0850. find
  15. Alex Graves and Jürgen Schmidhuber. 2005. Frame-wise Phoneme Classification with Bidirectional LSTM and Other Neural Network Architectures. Neural Networks, 18(5):602–610. find
  16. Matthew W Hartley and David E Reich. 2005. Method and System for Speech Recognition Using Phonetically Similar Word Alternatives. US Patent 6,910,012. find
  17. Chuen-Min Huang, Mei-Chen Wu, and Ching-Che Chang. 2007. Error Detection and Correction Based on Chinese Phonemic Alphabet in Chinese Text. In International Conference on Modeling Decisions for Artificial Intelligence, pages 463–476. Springer. find
  18. Chuen-Min Huang, Mei-Chen Wu, and Ching-Che Chang. 2008. Error Detection and Correction Based on Chinese Phonemic Alphabet in Chinese Text. International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, 16(supp01):89–105. find
  19. Frederick Jelinek. 1997. Statistical Methods for Speech Recognition. MIT press. find
  20. Wenbin Jiang, Liang Huang, Qun Liu, and Yajuan Lü. 2008. A Cascaded Linear Model for Joint Chinese Word Segmentation and Part-of-speech Tagging. Proceedings of ACL-08: HLT, pages 897–904. find
  21. Satoshi Kaki, Eiichiro Sumita, and Hitoshi Iida. 1998. A Method for Correcting Errors in Speech Recognition Using the Statistical Features of Character Co-occurrence. In Proceedings of the 36th Annual Meeting of the Association for Computational Linguistics and 17th International Conference on Compu… find
  22. Lung-Hao Lee, RAO Gaoqi, Liang-Chih Yu, XUN Endong, Baolin Zhang, and Li-Ping Chang. 2016. Overview of nlp-tea 2016 Shared task for Chinese Grammatical Error Diagnosis. In Proceedings of the 3rd Workshop on Natural Language Processing Techniques for Educational Applications (NLPTEA2016), pages 40–48… find
  23. C-L Liu, M-H Lai, K-W Tien, Y-H Chuang, S-H Wu, and C-Y Lee. 2011. Visually and Phonologically Similar Characters in Incorrect Chinese Words: Analyses, Identification, and Applications. ACM Transactions on Asian Language Information Processing (TALIP), 10(2):10. find
  24. Chao-Lin Liu, Kan-Wen Tien, Min-Hua Lai, Yi-Hsuan Chuang, and Shih-Hung Wu. 2009. Phonological and Logographic Influences on Errors in Written Chinese Words. In Proceedings of the 7th Workshop on Asian Language Resources, pages 84–91. Association for Computational Linguistics. find
  25. Deryle Lonsdale and Diane Strong-Krause. 2003. Automated Rating of ESL Essays. In Proceedings of the HLT-NAACL 03 workshop on Building educational applications using natural language processing-Volume 2, pages 61–67. Association for Computational Linguistics. find
  26. Bruno Martins and Mário J Silva. 2004. Spelling correction for search engine queries. In Advances in Natural Language Processing, pages 372–383. Springer. find
  27. Catherine McBride-Chang, Hua Shu, Aibao Zhou, Chun Pong Wat, and Richard K Wagner. 2003. Morphological Awareness Uniquely Predicts Young Children’s Chinese Character Recognition. Journal of educational psychology, 95(4):743. find
  28. G Nagy. 1988. Chinese Character Recognition: A Twenty-five-year Retrospective. In Pattern Recognition, 1988., 9th International Conference on, pages 163–167. IEEE. find
  29. Daniel Povey, Arnab Ghoshal, Gilles Boulianne, Lukas Burget, Ondrej Glembek, Nagendra Goel, Mirko Hannemann, Petr Motlicek, Yanmin Qian, Petr Schwarz, et al. 2011. find
  30. The Kaldi Speech Recognition Toolkit. In IEEE 2011 workshop on automatic speech recognition and understanding, EPFL-CONF-192584. IEEE Signal Processing Society. find
  31. Ray Smith. 2007. An Overview of the Tesseract OCR Engine. In Document Analysis and Recognition, 2007. find
  32. ICDAR 2007. Ninth International Conference on, volume 2, pages 629–633. IEEE. find
  33. Yan Song and Fei Xia. 2012. Using a Goodness Measurement for Domain Adaptation: A Case Study on Chinese Word Segmentation. In Proceedings of the Eighth International Conference on Language Resources and Evaluation (LREC-2012), pages 3853– 3860, Istanbul, Turkey. find
  34. Weiwei Sun. 2011. A Stacked Sub-Word Model for Joint Chinese Word Segmentation and Part-of-speech Tagging. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies-Volume 1, pages 1385–1394. Association for Computational Linguistics. find
  35. Shaohua Yang, Hai Zhao, Xiaolin Wang, and Bao-liang Lu. 2012. Spell Checking for Chinese. In LREC, pages 730–736. find
  36. Junjie Yu and Zhenghua Li. 2014. Chinese Spelling Error Detection and Correction Based on Language Model, Pronunciation, and shape. find
  37. Liang-Chih Yu, Lung-Hao Lee, Yuen-Hsien Tseng, Hsin-Hsi Chen, et al. 2014. Overview of SIGHAN 2014 Bake-off for Chinese Spelling Check. In Proceedings of the 3rd CIPSSIGHAN Joint Conference on Chinese Language Processing (CLP’14), pages 126–132. find
  38. Bo Zheng, Wanxiang Che, Jiang Guo, and Ting Liu. 2016. Chinese Grammatical Error Diagnosis with Long Short-Term Memory Networks. find
  39. Xiang Tong and David A Evans. 1996. A Statistical Approach to Automatic OCR Error Correction in Context. In Fourth Workshop on Very Large Corpora. find
  40. Yuen-Hsien Tseng, Lung-Hao Lee, Li-Ping Chang, and Hsin-Hsi Chen. 2015. Introduction to Sighan 2015 Bake-off for Chinese Spelling Check. find
  41. Kimberly Voll, Stella Atkins, and Bruce Forster. 2008. Improving the Utility of Speech Recognition Through Error Detection. Journal of digital imaging, 21(4):371. find
  42. Yih-Ru Wang and Yuan-Fu Liao. 2015. Word vector/conditional random field-based chinese spelling error detection for sighan-2015 evaluation. In Proceedings of the Eighth SIGHAN Workshop on Chinese Language Processing, pages 46–49. find
  43. Yih-Ru Wang, Yuan-Fu Liao, Yeh-Kuang Wu, and Liang-Chun Chang. 2013. Conditional Random Field-based Parser and Language Model for Traditional Chinese Spelling Checker. In Proceedings of the Seventh SIGHAN Workshop on Chinese Language Processing, pages 69–73. find
  44. Shih-Hung Wu, Yong-Zhi Chen, Ping-Che Yang, Tsun Ku, and Chao-Lin Liu. 2010. Reducing the False Alarm Rate of Chinese Character Error Detection and Correction. In CIPS-SIGHAN Joint Conference on Chinese Language Processing. find
  45. Shih-Hung Wu, Chao-Lin Liu, and Lung-Hao Lee. 2013. Chinese Spelling Check Evaluation at SIGHAN Bake-off 2013. find
  46. Nianwen Xue. 2003. Chinese Word Segmentation as Character Tagging. International Journal of Computational Linguistics & Chinese Language Processing, Volume 8, Number 1, February 2003: Special Issue on Word Formation and Chinese Language Processing, 8(1):29–48.

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/wang-2018-hybrid-approach-automatic-corpus-generation/