Elamur Research⌕
☰
home / papers / 2013 / Exploiting Zero Pronouns to Improve Chinese Coreference Resolution

Exploiting Zero Pronouns to Improve Chinese Coreference Resolution

Summary

Chinese often leaves pronouns unexpressed. Kong and Ng ask whether recovering these zero pronouns (ZPs) can help a Chinese coreference resolver. They start from a learning-based baseline that has two parts: an anaphoricity classifier and a mention-pair resolver. The resolver uses the 12 features of Soon et al. (2001) plus extra features for pronoun type, speakers and document parts, all trained with SVM-light on the CoNLL-2012 shared task data. To find ZPs, the authors detect clauses with a simplified semantic role labeling framework. They then use the detected ZPs in two ways. First, they retrain the syntactic parser with ZPs inserted. Second, they change how training examples are generated, treating ZPs as a special pronoun. With automatically detected ZPs, the system improves under every mention setting. With automatic mentions and with gold mention boundaries, it beats the top three CoNLL-2012 Chinese closed-track systems. This matters because it is reported as the first work to show that ZPs give a significant gain for Chinese coreference.

Key points

  • In a corpus study of OntoNotes v5.0, anaphoric ZPs make up 10.7% of coreference-chain mentions in the training data and 11.3% in the development data.
  • Adding ZPs to the Berkeley parser improves parsing F-measure by 1.8% with gold ZPs and by 1.4% with automatically detected ZPs.
  • With automatic mentions, the average F-measure rises from 60.71 to 63.07. MUC gains 3.31, BCUBED 1.42 and CEAF 2.34.
  • The refined parser mainly raises recall, while refined example generation mainly raises precision. Combining the two works best, and gold ZPs push the average to 65.14.
  • Automatic ZPs close roughly half of the gap between using no ZPs and using gold ZPs. For correctly identified anaphoric ZPs on the dev set, resolution precision reaches 94.76%.

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

Fields: Linguistics & languages

Topics: Coreference resolution

Cited by

9 papers of our library cite it (of the 140673 whose bibliographies we read); 3 of them have a page here, 6 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work.

@inproceedings{kong2013exploiting,
  title = {{Exploiting Zero Pronouns to Improve Chinese Coreference Resolution}},
  author = {Kong, Fang and Ng, Hwee Tou},
  booktitle = {Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing},
  year = {2013},
  url = {https://aclanthology.org/D13-1028/}
}

References

23 references, 14 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Anders Björkelund and Richárd Farkas. 2012. Data-driven multilingual coreference resolution using resolver stacking. In Proceedings of the Joint Conference on EMNLP and CoNLL – Shared Task, pages 49– 55. find our card
  2. Chen Chen and Vincent Ng. 2012. Combining the best of two worlds: A hybrid approach to multilingual coreference resolution. In Proceedings of the Joint Conference on EMNLP and CoNLL – Shared Task, pages 56–63. find our card
  3. Tagyoung Chung and Daniel Gildea. 2010. Effects of empty categories on machine translation. In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, pages 636–645. find
  4. Susan Converse. 2006. Pronominal Anaphora Resolution in Chinese. Ph.D. thesis, University of Pennsylvania. find
  5. Eraldo Rezende Fernandes, Cı́cero Nogueira dos Santos, and Ruy Luiz Milidiú. 2012. Latent structure perceptron with feature induction for unrestricted coreference resolution. In Proceedings of the Joint Conference on EMNLP and CoNLL – Shared Task, pages 41–48. find our card
  6. Thorsten Joachims. 1999. Making large-scale SVM learning practical. In Bernhard Schölkopf, Christopher J. C. Burges, and Alexander J. Smola, editors, Advances in Kernel Methods: Support Vector Learning. MIT-Press. find
  7. Young-Joo Kim. 2000. Subject/object drop in the acquisition of Korean: A cross-linguistic comparison. Journal of East Asian Linguistics, 9:325–351. find
  8. BCUBED P F 80.38 76.73 78.97 78.40 83.45 78.24 80.38 78.15 82.94 80.81 CEAF P 45.66 49.81 47.44 49.19 50.63 Avg F 56.27 58.07 56.98 59.58 62.28 R 73.40 77.83 73.64 76.04 78.79 R 53.16 50.40 51.12 53.98 54.12 F 49.13 50.10 49.21 51.47 52.32 60.71 62.19 61.48 63.07 65.14 find
  9. Fang Kong and Guodong Zhou. 2010. A tree kernel-based unified framework for Chinese zero anaphora resolution. In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, pages 882–891. find our card
  10. Fang Kong, Guodong Zhou, and Qiaoming Zhu. 2009. Employing the centering theory in pronoun resolution from the semantic perspective. In Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing, pages 987–996. find our card
  11. Heeyoung Lee, Yves Peirsman, Angel Chang, Nathanael Chambers, Mihai Surdeanu, and Dan Jurafsky. 2011. Stanford’s multi-pass sieve coreference resolution system at the CoNLL-2011 shared task. In Proceedings of the Fifteenth Conference on Computational Natural Language Learning: Shared Task, pages 28–… find our card
  12. Vincent Ng and Claire Cardie. 2002. Improving machine learning approaches to coreference resolution. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, pages 104–111. find our card
  13. Vincent Ng. 2007. Semantic class induction and coreference resolution. In Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics, pages 536–543. find our card
  14. Slav Petrov, Leon Barrett, Romain Thibaux, and Dan Klein. 2006. Learning accurate, compact, and interpretable tree annotation. In Proceedings of the 21st International Conference on Computational Linguistics and 44th Annual Meeting of the Association for Computational Linguistics, pages 433–440. find our card
  15. Simone Paolo Ponzetto and Michael Strube. 2006. Exploiting semantic role labeling, WordNet and Wikipedia for coreference resolution. In Proceedings of the Human Language Technology Conference of the NAACL, pages 192–199. find our card
  16. Wee Meng Soon, Hwee Tou Ng, and Daniel Chung Yong Lim. 2001. A machine learning approach to coreference resolution of noun phrases. Computational Linguistics, 27(4):521–544. find our card
  17. Yannick Versley. 2007. Antecedent selection techniques for high-recall coreference resolution. In Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, pages 496–505. find our card
  18. Bonnie Lynn Webber. 1978. A Formal Approach to Discourse Anaphora. Garland Press. find
  19. Nianwen Xue. 2008. Labeling Chinese predicates with semantic roles. Computational Linguistics, 34(2):225–255. find
  20. Yaqin Yang and Nianwen Xue. 2010. Chasing the ghost: Recovering empty categories in the Chinese Treebank. In Coling 2010: Posters, pages 1382–1390. find
  21. Bo Yuan, Qingcai Chen, Yang Xiang, Xiaolong Wang, Liping Ge, Zengjian Liu, Meng Liao, and Xianbo Si. 2012. A mixed deterministic model for coreference resolution. In Proceedings of the Joint Conference on EMNLP and CoNLL – Shared Task, pages 76–82. find our card
  22. Shanheng Zhao and Hwee Tou Ng. 2007. Identification and resolution of Chinese zero pronouns: A machine learning approach. In Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, pages 541–550. find our card
  23. Zhi Zhong and Hwee Tou Ng. 2010. It Makes Sense: A wide-coverage word sense disambiguation system for free text. In Proceedings of the ACL 2010 System Demonstrations, pages 78–83. find

↑ back to top