Elamur Research⌕
☰
home / papers / 2014 / SinoCoreferencer: An End-to-End Chinese Event Coreference Resolver

SinoCoreferencer: An End-to-End Chinese Event Coreference Resolver

Summary

Event coreference resolution groups event mentions in a text that describe the same real-world happening. It sits at the very end of an information extraction pipeline, and almost all prior systems target English. The authors present SinoCoreferencer, an end-to-end resolver for Chinese that follows the ACE task definitions and is released publicly. Raw text is first preprocessed with Stanford CoreNLP. It then flows through eight modules grouped into four subsystems: entity extraction, entity coreference, event extraction and event coreference. Each module can also be run on its own. The modules use character-level CRF sequence labelers (CRF++), SVM multiclass classifiers and a rule-based multi-pass sieve resolver for entities. Evaluated on the ACE 2005 Chinese data, the system reports state-of-the-art event coreference, scoring 73.4 average F with gold inputs and 49.9 with fully predicted inputs. Comparing these two settings shows how much errors accumulate along the pipeline. The authors believe it is the first publicly available Chinese event coreference tool, and its reusable components can support other Chinese NLP applications.

Key points

  • The architecture has eight components in four subsystems. Event coreference depends on the outputs of the event extraction and entity coreference subsystems, and both of those depend on entity extraction.
  • Entity mentions, time expressions and value expressions are found with character-based BIO tagging in CRF++. The features include character n-grams, 10 Chinese wordlists and POS/NP cues, and this step reaches 84.7% F.
  • Entity typing covers the 7 ACE 2005 types and 45 subtypes. The named entity tagger is trained on CoNLL-2012 Chinese data with 18 NE types (37 labels) and scores 66.4% F.
  • Entity coreference uses nine ordered sieves, partly modeled on Lee et al. (2011). It achieves 70.6 Avg F (MUC/B³/CEAFe) with gold input and 63.0 with predicted input.
  • Event coreference scores 73.4 Avg F with perfect input but 49.9 with predicted input. Argument identification drops even more sharply, from 76.9% to 28.3% F.

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); 4 of them have a page here, 5 not yet.

Cite

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

@inproceedings{chen2014sinocoreferencer,
  title = {{SinoCoreferencer: An End-to-End Chinese Event Coreference Resolver}},
  author = {Chen, Chen and Ng, Vincent},
  booktitle = {Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14)},
  year = {2014},
  doi = {10.63317/57n7frh2boyy},
  url = {https://doi.org/10.63317/57n7frh2boyy}
}

References

18 references, 13 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Bagga, A. and Baldwin, B. (1998). Algorithms for scoring coreference chains. In Proceedings of the Linguistic Coreference Workshop at The First International Conference on Language Resources and Evaluation, page 563--566. find our card
  2. Bejan, C. and Harabagiu, S. (2010). Unsupervised event coreference resolution with rich linguistic features. In Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics, pages 1412--1422. find our card
  3. Chen, Z. and Ji, H. (2009). Graph-based event coreference resolution. In Proceedings of the 2009 Workshop on Graph-based Methods for Natural Language Processing, pages 54--57. find our card
  4. Chen, C. and Ng, V. (2012). Joint modeling for Chinese event extraction with rich linguistic features. In Proceedings of the 24th International Conference on Computational Linguistics, pages 529--544. find
  5. Chen, Z., Ji, H., and Haralick, R. (2009). A pairwise event coreference model, feature impact and evaluation for event coreference resolution. In Proceedings of the RANLP 2009 Workshop on Events in Emerging Text Types, pages 17--22. find our card
  6. Chen, B., Su, J. and Tan, C. L. (2010). Resolving Event Noun Phrases to Their Verbal Mentions. In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, pages 872--881. find
  7. Chen, B., Su, J., Pan, S. J., and Tan, C. L. (2011). A unified event coreference resolution by integrating multiple resolvers. In Proceedings of 5th International Joint Conference on Natural Language Processing, pages 102-- 110. find our card
  8. Haghighi, A. and Klein, D. (2009). Simple coreference resolution with rich syntactic and semantic features. In Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing, pages 1152--1161. find our card
  9. Humphreys, K., Gaizauskas, R., and Azzam, S. (1997). Event coreference for information extraction. In Proceedings of the Workshop on Operational Factors in Practical, Robust Anaphora Resolution for Unrestricted Texts, pages 75--81. find our card
  10. Joachims, T. (1999). Making large-scale SVM learning practical. In Scholkopf, B. and Smola, A., editors, Advances in Kernel Methods - Support Vector Learning, pages 44--56. MIT Press. find
  11. Lee, H., Peirsman, Y., Chang, A., Chambers, N., Surdeanu, M., and Jurafsky, D. (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--34. find our card
  12. Lee, H., Recasens, M., Chang, A., Surdeanu, M., and Jurafsky, D. (2012). Joint entity and event coreference resolution across documents. In Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, pages 489--500. find our card
  13. Luo, X. (2005). On coreference resolution performance metrics. In Proceedings of Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing, pages 25--32. find our card
  14. Mitkov, R., Boguraev, B., and Lappin, S. (2001). Introduction to the special issue on computational anaphora resolution. Computational Linguistics, 27(4):473--477. find
  15. Raghunathan, K., Lee, H., Rangarajan, S., Chambers, N., Surdeanu, M., Jurafsky, D., and Manning, C. (2010). A multi-pass sieve for coreference resolution. In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, pages 492--501. find our card
  16. Soon, W. M., Ng, H. T., and Lim, D. C. Y. (2001). A machine learning approach to coreference resolution of noun phrases. Computational Linguistics, 27(4):521--544. find our card
  17. Tsochantaridis, I., Hofmann, T., Joachims, T., and Altun, Y. (2004). Support vector machine learning for interdependent and structured output spaces. In Proceedings of the 21st International Conference on Machine Learning, pages 104--112. find
  18. Vilain, M., Burger, J., Aberdeen, J., Connolly, D., and Hirschman, L. (1995). A model-theoretic coreference scoring scheme. In Proceedings of the Sixth Message Understanding Conference, pages 45--52. find our card

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/chen-2014-sinocoreferencer-end-end-chinese-event/