Elamur Research⌕
☰
home / papers / 2013 / Chinese Zero Pronoun Resolution: Some Recent Advances

Chinese Zero Pronoun Resolution: Some Recent Advances

Summary

Chinese often leaves subjects unexpressed. These gaps, called zero pronouns, are hard to link to earlier noun phrases because they carry no number or gender cues. The authors build on the Zhao and Ng (2007) mention-pair resolver for anaphoric zero pronouns (AZPs), which uses SVM light, and add two extensions:

  1. Contextual features:
    • lexical pairings of a candidate antecedent's head with the verb (and object) after the gap, as a rough stand-in for selectional preferences;
    • a simplified narrative-event-chain cue based on identical governing verbs;
    • the sentence-final punctuation mark.
  2. Links between zero pronouns: already-resolved AZPs become candidate antecedents. These links act as stepping stones toward distant overt antecedents.

Experiments use the CoNLL-2012 portion of OntoNotes 4.0. The extended system beats both the Zhao and Ng baseline and the tree-kernel resolver of Kong and Zhou (2010) in three settings: gold input, system-detected AZPs, and fully automatic parses. The paper is also the first to report end-to-end Chinese zero pronoun resolution, giving a realistic benchmark beyond results that depend on gold annotations.

Key points

  • Data: CoNLL-2012 OntoNotes 4.0. Only the training and development sets carry ZP annotation, so the development set serves as the test set.
    • Training: 1,391 documents, 12,111 AZPs.
    • Test: 172 documents, 1,713 AZPs.
  • Gold AZPs and gold parse trees: F-score rises from 41.5 (Z&N baseline) to 47.7 (full system). K&Z scores 38.0. The new features alone add 4.7 points; the ZP links alone add 1.2.
  • End-to-end setting (system AZPs, automatic parses): F-score improves from 13.4 to 15.7. K&Z reaches 11.9.
  • AZP identification: an SVM classifier with 32 features reaches 52.8 F on gold parses but only 32.5 F on automatic parses.
  • ZP links: they help little with system-detected AZPs, because identification errors introduce many incorrect links.

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

19 papers of our library cite it (of the 140673 whose bibliographies we read); 13 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{chen2013chinese,
  title = {{Chinese Zero Pronoun Resolution: Some Recent Advances}},
  author = {Chen, Chen and Ng, Vincent},
  booktitle = {Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing},
  year = {2013},
  doi = {10.18653/v1/d13-1135},
  url = {https://doi.org/10.18653/v1/d13-1135}
}

References

10 references, 5 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Nathanael Chambers and Dan Jurafsky. 2008. Unsupervised learning of narrative event chains. In Proceedings of the 46th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, pages 787--797. find
  2. Thorsten Joachims. 1999. Making large-scale SVM learning practical. In Bernhard Scholkopf and Alexander Smola, editors, Advances in Kernel Methods - Support Vector Learning, pages 44--56. MIT Press. find
  3. Andrew Kehler, Douglas Appelt, Lara Taylor, and Aleksandr Simma. 2004. Competitive self-trained pronoun interpretation. In Proceedings of HLT-NAACL 2004: Short Papers, pages 33--36. find
  4. 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
  5. Alessandro Moschitti. 2006. Making tree kernels practical for natural language processing. In Proceedings of the 11th Conference of the European Chapter of the Association for Computational Linguistics, pages 113--120. find
  6. 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
  7. Yaqin Yang and Nianwen Xue. 2010. Chasing the ghost: recovering empty categories in the Chinese Treebank. In Proceedings of the 23rd International Conference on Computational Linguistics: Posters, pages 1382-- 1390. find
  8. Xiaofeng Yang, Jian Su, and Chew Lim Tan. 2005. Improving pronoun resolution using statistics-based semantic compatibility information. In Proceedings of the 43rd Annual Meeting of the Association for Computational Linguistics, pages 165--172. find our card
  9. 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 on Natural Language Processing and Computational Natural Language Learning, pages 541--550. find our card
  10. GuoDong Zhou, Fang Kong, and Qiaoming Zhu. 2008. Context-sensitive convolution tree kernel for pronoun resolution. In Proceedings of the 3rd International Joint Conference on Natural Language Processing, pages 25--31. find our card

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/chen-2013-chinese-zero-pronoun-resolution-some/