Elamur Research⌕
☰
home / papers / 2011 / A Unified Event Coreference Resolution by Integrating Multiple Resolvers

A Unified Event Coreference Resolution by Integrating Multiple Resolvers

Summary

Most coreference work targets objects such as people or places. This paper focuses on event coreference, where verbs, noun phrases and pronouns can all point to the same eventuality. The task is harder than object coreference because events are distinguished by their arguments (time, place, participants), and standard coreference features work poorly when mentions belong to different syntactic categories. Working on OntoNotes 2.0, the authors build a two-step system:

  1. Seven mention-pair resolvers, three of them new, cover both event and object cases. They use SVMs with a composite kernel that combines flat features with a convolution tree kernel over the minimum-expansion parse path.
  2. Chains are then formed with spectral graph partitioning.

Three enhancements are added on top:

  • Object resolvers' outputs are used to reject spurious event links.
  • Training instances are reselected so that classifiers predict links between any two mentions, not just local candidate preferences.
  • Coreferent pronoun information is fed into the partitioning step.

Together these raise event-chain B3 F-score from 38.33% to 46.91%. The work matters because event extraction systems need such chains to gather event details scattered across sentences.

Key points

  • In OntoNotes 2.0, 68.05% of documents have at least one event chain, and event chains make up 15.52% of all coreference chains.
  • Using competing object-coreference classifiers improves the Verb-Pronoun resolver by more than 9 F points and the event NP-Pronoun resolver by about 7. Chain-level B3 F rises from 38.33% to 43.61%.
  • Revised instance selection has mixed effects at the pair level but helps chains:
    • Best-candidate F falls by 3.26 points.
    • Coreferent-link F rises by 2.84 points.
    • B3 chain F improves by 2.02 points (to 45.63%).
  • Adding pronoun information to spectral partitioning gives +2.19 F (45.82%). Both this gain and the instance-selection gain are significant under a paired t-test at the 5% level.
  • The best configuration (BL+CC+RIS+Pron) reaches B3 P/R/F of 35.27/70.02/46.91%, an overall gain of 8.54 points.

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

16 papers of our library cite it (of the 140673 whose bibliographies we read); 7 of them have a page here, 9 not yet.

Cite

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

@inproceedings{chen2011unified,
  title = {{A Unified Event Coreference Resolution by Integrating Multiple Resolvers}},
  author = {Chen, Bin and Su, Jian and Pan, Sinno Jialin and Tan, Chew Lim},
  booktitle = {Proceedings of 5th International Joint Conference on Natural Language Processing},
  year = {2011},
  url = {https://aclanthology.org/I11-1012/}
}

References

17 references, 9 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Asher,N. 1993. Reference to Abstract Objects in Discourse. Kluwer Academic Publisher. find
  2. Bagga,A. & Baldwin,B. 1998. Algorithms for scoring coreference chains. In Proceedings of the Linguistic Coreference Workshop at Conference on Language Resources and Evaluation (LREC-1998). find our card
  3. Byron,D. 2002. Resolving Pronominal Reference to Abstract Entities, In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics (ACL), USA. find
  4. Cai,J. & Strube,M. 2010. End-to-End Coreference Resolution via Hypergraph Partitioning. In Proceedings of the International Conference on Computational Linguistics (COLING), China, Chen,B.; Su,J. & Tan,C.L. 2010a. A Twin-Candidate Based Approach for Event Pronoun Resolution using Composite Kernel. I… find our card
  5. Hovy,E.; Marcus,M.; Palmer,M.; Ramshaw,L. & Weischedel,R. 2006. OntoNotes: The 90% Solution. In Proceedings of the Human Language Technology Conference of the NAACL, USA Luxburg,U. 2006. A tutorial on spectral clustering. In (MPITechnical Reports No. 149). Tubingen: Max Planck Institute for Biologic… find
  6. Ng,A.; Jordan,M. & Weiss,Y. 2002. On spectral clustering: analysis and an algorithm. In Advances in Neural Information Processing Systems 14 (pp. 849 – 856). MIT Press Ng,V. & Cardie,C. 2002. Combining sample selection and error-driven pruning for machine learning of coreference rules. In Proceeding… find our card
  7. Nicolae,C & Nicolae,G. 2006. BESTCUT: A Graph Algorithm for Coreference Resolution. In Proceedings of conference on Empirical Methods in Natural Language Processing (EMNLP), Austrilia. find our card
  8. Poon,H. & Domingos.P. 2008. Joint Unsupervised Coreference Resolution with Markov Logic. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), 2008. find our card
  9. Pradhan,S.; Ramshaw,L.; Weischedel,R.; MacBride,J. & Micciulla,L. 2007. Unrestricted Coreference: Identifying Entities and Events in OntoNotes. In Proceedings of the IEEE International Conference on Semantic Computing (ICSC), USA. find
  10. Shamir,R.; and Sharan,R. 2002. Algorithmic approaches to clustering gene expression data. Current Topics in Computational Molecular Biology, 269–300. find
  11. Shi,J. & Malik,J. 2000. Normalized cuts and image segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence(PAMI). find
  12. Sidner,C.L. 1983. Focusing in the comprehension of definite anaphora. In Computational Models of Discourse. MIT Press. find
  13. Soon,W.; Ng,H. & Lim, D. 2001. A machine learning approach to coreference resolution of noun phrases. Computational Linguistics, 27(4):521– 544. find our card
  14. Stoyanov,V.; Cardie,C.; Gilbert,N.; Riloff,E.; Buttler,D. & Hysom,D. 2010 Coreference Resolution with Reconcile. In Proceedings of the Conference of the 48th Annual Meeting of the Association for Computational Linguistics (ACL 2010) find our card
  15. Versley,Y.; Ponzetto,S.P.; Poesio,M.; Eidelman,V.; Jern,A.; Smith,J.; Yang,X. & Moschitti,A. 2008. BART: A Modular Toolkit for Coreference Resolution. In Proceedings of the 6th International Conference on Language Resources and Evaluation (LREC 2008). find our card
  16. Yang,X.; Su,J. & Tan,C.L. 2006. Kernel-Based Pronoun Resolution with Structured Syntactic Knowledge. In Proceedings of the Conference of the 46th Annual Meeting of the Association for Computational Linguistics. Australia. find our card
  17. Yang,X.; Su,J. & Tan,C.L. 2008. A Twin-Candidates Model for Learning-Based Coreference Resolution. In Computational Linguistics, 34(3):327-356. find

↑ back to top

This page in Markdown →