Elamur Research⌕
☰
home / papers / 2013 / Multigraph Clustering for Unsupervised Coreference Resolution

Multigraph Clustering for Unsupervised Coreference Resolution

Summary

This paper addresses coreference resolution, the task of grouping text mentions that point to the same entity, without any training data. Each document becomes a directed labeled weighted multigraph. Mentions are the nodes, and edges carry pairwise relations that may link a later mention to an earlier one. The relations are standard indicators taken from earlier work. Negative constraints, such as gender or number disagreement, receive a weight of −∞. Positive cues, such as string match or syntactic parallelism, each receive a weight of 1. Because no weights are learned, the model counts as unsupervised. Clustering is done greedily with a 1-nearest-neighbor strategy: each mention attaches to the earlier candidate with the most positive relations and no negative one, with ties going to the closest candidate. On the English track of the CoNLL'12 shared task, under closed-track conditions, the system beats most participating systems. It also compares favorably with supervised weighting variants. An error analysis points to ambiguous pronouns and nominals with different heads as the main weaknesses. The results suggest that simple, carefully chosen relations can rival heavily engineered or learned systems.

Key points

  • The relation set has 23 binary relations: 9 negative (e.g., N_Gender, N_Number, N_Mod, N_ItDist) and 14 positive (e.g., P_AnaPron, P_Subject, P_Pron_StrMatch). Relations and their distance thresholds were tuned on development data.
  • The data is the CoNLL'12 English data, a subset of OntoNotes 5.0, with the official dev/test split. Resources are limited to WordNet, Bergsma and Lin's (2006) number/gender data, and the provided annotation layers. System mentions come from the extractor of Martschat et al. (2012).
  • The unsupervised model's F1 gains over the supervised weighting variants are statistically significant by paired t-test (p < 0.05). The one exception is the MUC F1 comparison against the "weights fraction" variant.
  • Precision is weakest on pronouns such as it, with 64% of 1,114 decisions wrong, and on nominal–nominal pairs, at 49% wrong. Speaker relations make I and my comparatively reliable.
  • The development set contains 3,528 recall errors, mostly nominal–nominal pairs. In a sample of 50 such errors, 29 stemmed from missing semantic knowledge, 14 from an overly strict N_Mod, and 4 from overly cautious string matching.

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

7 papers of our library cite it (of the 140673 whose bibliographies we read); 2 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{martschat2013multigraph,
  title = {{Multigraph Clustering for Unsupervised Coreference Resolution}},
  author = {Martschat, Sebastian},
  booktitle = {51st Annual Meeting of the Association for Computational Linguistics Proceedings of the Student Research Workshop},
  year = {2013},
  url = {https://aclanthology.org/P13-3012/}
}

References

28 references, 24 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Roxana Angheluta, Patrick Jeuniaux, Rudradeb Mitra, and Marie-Francine Moens. 2004. Clustering algorithms for noun phrase coreference resolution. In Proceedings of the 7èmes Journées Internationales d’Analyse Statistique des Données Textuelles, Louvain La Neuve, Belgium, 10–12 March 2004, pages 6… find
  2. Amit Bagga and Breck Baldwin. 1998. Algorithms for scoring coreference chains. In Proceedings of the 1st International Conference on Language Resources and Evaluation, Granada, Spain, 28–30 May 1998, pages 563–566. find our card
  3. Shane Bergsma and Dekang Lin. 2006. Bootstrapping path-based pronoun resolution. In Proceedings of the 21st International Conference on Computational Linguistics and 44th Annual Meeting of the Association for Computational Linguistics, Sydney, Australia, 17–21 July 2006, pages 33–40. find our card
  4. Jie Cai and Michael Strube. 2010. End-to-end coreference resolution via hypergraph partitioning. In Proceedings of the 23rd International Conference on Computational Linguistics, Beijing, China, 23– 27 August 2010, pages 143–151. find our card
  5. Jie Cai, Éva Mújdricza-Maydt, Yufang Hou, and Michael Strube. 2011a. Weakly supervised graph-based coreference resolution for clinical data. In Proceedings of the 5th i2b2 Shared Tasks and Workshop on Challenges in Natural Language Processing for Clinical Data, Washington, D.C., 20-21 October 2011… find
  6. Jie Cai, Éva Mújdricza-Maydt, and Michael Strube. 2011b. Unrestricted coreference resolution via global hypergraph partitioning. In Proceedings of the Shared Task of the 15th Conference on Computational Natural Language Learning, Portland, Oreg., 23–24 June 2011, pages 56–60. find our card
  7. Claire Cardie and Kiri Wagstaff. 1999. Noun phrase coreference as clustering. In Proceedings of the 1999 SIGDAT Conference on Empirical Methods in Natural Language Processing and Very Large Corpora, College Park, Md., 21–22 June 1999, pages 82–89. find our card
  8. Eugene Charniak and Micha Elsner. 2009. EM works for pronoun anaphora resolution. In Proceedings of the 12th Conference of the European Chapter of the Association for Computational Linguistics, Athens, Greece, 30 March – 3 April 2009, pages 148–156. find our card
  9. Aron Culotta, Michael Wick, and Andrew McCallum. 2007. First-order probabilistic models for coreference resolution. In Proceedings of Human Language Technologies 2007: The Conference of the North American Chapter of the Association for Computational Linguistics, Rochester, N.Y., 22–27 April 2007, pa… find our card
  10. Christiane Fellbaum, editor. 1998. WordNet: An Electronic Lexical Database. MIT Press, Cambridge, Mass. find
  11. Eraldo Fernandes, Cı́cero dos Santos, and Ruy Milidiú. 2012. Latent structure perceptron with feature induction for unrestricted coreference resolution. In Proceedings of the Shared Task of the 16th Conference on Computational Natural Language Learning, Jeju Island, Korea, 12–14 July 2012, pages 41… find our card
  12. Aria Haghighi and Dan Klein. 2007. Unsupervised coreference resolution in a nonparametric Bayesian model. In Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics, Prague, Czech Republic, 23–30 June 2007, pages 848–855. find our card
  13. 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 Shared Task of the 15th Conference on Computational Natural Language Learning, Portland… find our card
  14. Xiaoqiang Luo. 2005. On coreference resolution performance metrics. In Proceedings of the Human Language Technology Conference and the 2005 Conference on Empirical Methods in Natural Language Processing, Vancouver, B.C., Canada, 6–8 October 2005, pages 25–32. find our card
  15. Sebastian Martschat, Jie Cai, Samuel Broscheit, Éva Mújdricza-Maydt, and Michael Strube. 2012. A multigraph model for coreference resolution. In Proceedings of the Shared Task of the 16th Conference on Computational Natural Language Learning, Jeju Island, Korea, 12–14 July 2012, pages 100–106. find
  16. Ruslan Mitkov. 1998. Robust pronoun resolution with limited knowledge. In Proceedings of the 17th International Conference on Computational Linguistics and 36th Annual Meeting of the Association for Computational Linguistics, Montréal, Québec, Canada, 10–14 August 1998, pages 869–875. find our card
  17. Vincent Ng. 2008. Unsupervised models for coreference resolution. In Proceedings of the 2008 Conference on Empirical Methods in Natural Language Processing, Waikiki, Honolulu, Hawaii, 25–27 October 2008, pages 640–649. find our card
  18. Cristina Nicolae and Gabriel Nicolae. 2006. BestCut: A graph algorithm for coreference resolution. In Proceedings of the 2006 Conference on Empirical Methods in Natural Language Processing, Sydney, Australia, 22–23 July 2006, pages 275–283. find our card
  19. Hoifung Poon and Pedro Domingos. 2008. Joint unsupervised coreference resolution with Markov Logic. In Proceedings of the 2008 Conference on Empirical Methods in Natural Language Processing, Waikiki, Honolulu, Hawaii, 25–27 October 2008, pages 650– 659. find our card
  20. Sameer Pradhan, Lance Ramshaw, Mitchell Marcus, Martha Palmer, Ralph Weischedel, and Nianwen Xue. 2011. CoNLL-2011 Shared Task: Modeling unrestricted coreference in OntoNotes. In Proceedings of the Shared Task of the 15th Conference on Computational Natural Language Learning, Portland, Oreg., 23–24… find our card
  21. Sameer Pradhan, Alessandro Moschitti, and Nianwen Xue. 2012. CoNLL-2012 Shared Task: Modeling multilingual unrestricted coreference in OntoNotes. In Proceedings of the Shared Task of the 16th Conference on Computational Natural Language Learning, Jeju Island, Korea, 12–14 July 2012, pages 1– 40. find our card
  22. Karthik Raghunathan, Heeyoung Lee, Sudarshan Rangarajan, Nathanael Chambers, Mihai Surdeanu, Dan Jurafsky, and Christopher Manning. 2010. A multi-pass sieve for coreference resolution. In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, Cambridge, Mass., 9–11 O… find our card
  23. Altaf Rahman and Vincent Ng. 2009. Supervised models for coreference resolution. In Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing, Singapore, 6–7 August 2009, pages 968–977. find our card
  24. Emili Sapena, Lluı́s Padró, and Jordi Turmo. 2010. A global relaxation labeling approach to coreference resolution. In Proceedings of Coling 2010: Poster Volume, Beijing, China, 23–27 August 2010, pages 1086–1094. find our card
  25. Emili Sapena, Lluı́s Padró, and Jordi Turmo. 2011. RelaxCor participation in CoNLL shared task on coreference resolution. In Proceedings of the Shared Task of the 15th Conference on Computational Natural Language Learning, Portland, Oreg., 23–24 June 2011, pages 35–39. find our card
  26. 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
  27. Yannick Versley, Simone Paolo Ponzetto, Massimo Poesio, Vladimir Eidelman, Alan Jern, Jason Smith, Xiaofeng Yang, and Alessandro Moschitti. 2008. BART: A modular toolkit for coreference resolution. In Companion Volume to the Proceedings of the 46th Annual Meeting of the Association for Computational… find our card
  28. Marc Vilain, John Burger, John Aberdeen, Dennis Connolly, and Lynette Hirschman. 1995. A model-theoretic coreference scoring scheme. In Proceedings of the 6th Message Understanding Conference (MUC-6), pages 45–52, San Mateo, Cal. Morgan Kaufmann. find our card

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/martschat-2013-multigraph-clustering-unsupervised-coreference-resolution/