Elamur Research⌕
☰
home / papers / 2008 / An Entity-Mention Model for Coreference Resolution with Inductive Logic Programming

An Entity-Mention Model for Coreference Resolution with Inductive Logic Programming

Summary

Most learning-based coreference systems use the mention-pair model, which classifies two mentions at a time. This misses evidence spread across the rest of an entity: a pronoun alone, for example, reveals little beyond gender. Earlier entity-mention models addressed this with hand-crafted first-order features that summarize an entity, such as "Any-X" or "Most-X" aggregations. These features are ad hoc and cannot describe individual mentions inside the entity.

The authors recast entity-level coreference with Inductive Logic Programming (ILP). Entities, their member mentions and the active mention are encoded as relational facts, using a has_mention relation plus standard lexical, syntactic and agreement features. ILP then learns first-order rules that decide whether a mention should join a partially built entity. Resolution uses greedy clustering.

On the ACE data set (NWire, NPaper and BNews domains), the ILP entity-mention model outperforms:

  • an ILP mention-pair baseline, with F-measure gains of up to 2.3%;
  • an entity-mention model built on heuristic first-order features.

The work is presented as the first use of ILP for coreference. It shows that learned relational rules can exploit entity-wide information for this task.

Key points

  • Evaluation is on ACE across three domains: NWire, NPaper and BNews. The baselines are Soon et al. (2001)-style mention-pair systems trained with C4.5 or ILP.
  • Against the ILP mention-pair model, the entity-mention model's F-measure rises by 2.3% (NWire), 0.4% (NPaper) and 1.4% (BNews). The NWire and BNews gains are statistically significant (2-tailed t test, p < 0.05).
  • The gains come from precision, which rises by up to 4.6% (BNews). Recall falls by at most 1.8%.
  • The ILP model also beats the entity-mention model with manually designed "Any-X" features by 1.0–1.8% F-measure.
  • Adding all antecedents of an entity as training pairs in the mention-pair model hurts it: precision drops by up to 12%. Learned rules, by contrast, can refer to several mentions of an entity at once, for example a rule linking a pronoun to an entity that contains both a name and an indefinite NP in subject position.

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

29 papers of our library cite it (of the 140673 whose bibliographies we read); 16 of them have a page here, 13 not yet.

Cite

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

@inproceedings{yang2008entity,
  title = {{An Entity-Mention Model for Coreference Resolution with Inductive Logic Programming}},
  author = {Yang, Xiaofeng and Su, Jian and Lang, Jun and Tan, Chew Lim and Liu, Ting and Li, Sheng},
  booktitle = {Proceedings of ACL-08: HLT},
  year = {2008},
  url = {https://aclanthology.org/P08-1096/}
}

References

21 references, 11 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. C. Aone and S. W. Bennett. 1995. Evaluating automated and manual acquisition of anaphora resolution strategies. In Proceedings of the 33rd Annual Meeting of the Association for Computational Linguistics (ACL), pages 122–129. find our card
  2. V. Claveau, P. Sebillot, C. Fabre, and P. Bouillon. 2003. Learning semantic lexicons from a part-of-speech and semantically tagged corpus using inductive logic programming. Journal of Machine Learning Research, 4:493–525. find
  3. A. Culotta, M. Wick, and A. McCallum. 2007. First-order probabilistic models for coreference resolution. In Proceedings of the Annual Meeting of the North America Chapter of the Association for Computational Linguistics (NAACL), pages 81–88. find our card
  4. J. Cussens. 1996. Part-of-speech disambiguation using ilp. Technical report, Oxford University Computing Laboratory. find
  5. P. Denis and J. Baldridge. 2007. Joint determination of anaphoricity and coreference resolution using integer programming. In Proceedings of the Annual Meeting of the North America Chapter of the Association for Computational Linguistics (NAACL), pages 236–243. find our card
  6. X. Luo, A. Ittycheriah, H. Jing, N. Kambhatla, and S. Roukos. 2004. A mention-synchronous coreference resolution algorithm based on the bell tree. In Proceedings of the 42nd Annual Meeting of the Association for Computational Linguistics (ACL), pages 135–142. find our card
  7. A. McCallum and B. Wellner. 2003. Toward conditional models of identity uncertainty with application to proper noun coreference. In Proceedings of IJCAI-03 Workshop on Information Integration on the Web, pages 79–86. find
  8. J. McCarthy and W. Lehnert. 1995. Using decision trees for coreference resolution. In Proceedings of the 14th International Conference on Artificial Intelligences (IJCAI), pages 1050–1055. find
  9. R. Mooney. 1997. Inductive logic programming for natural language processing. In Proceedings of the sixth International Inductive Logic Programming Workshop, pages 3–24. find
  10. V. Ng and C. Cardie. 2002. Improving machine learning approaches to coreference resolution. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics (ACL), pages 104–111, Philadelphia. find our card
  11. V. Ng. 2005. Machine learning for coreference resolution: From local classification to global ranking. In Proceedings of the 43rd Annual Meeting of the Association for Computational Linguistics (ACL), pages 157–164. find our card
  12. V. Ng. 2007. Semantic class induction and coreference resolution. In Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics (ACL), pages 536–543. find our card
  13. W. Soon, H. Ng, and D. Lim. 2001. A machine learning approach to coreference resolution of noun phrases. Computational Linguistics, 27(4):521–544. find our card
  14. L. Specia, M. Stevenson, and M. V. Nunes. 2007. Learning expressive models for words sense disambiguation. In Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics (ACL), pages 41–48. find
  15. A. Srinivasan. 2000. The aleph manual. Technical report, Oxford University Computing Laboratory. find
  16. M. Vilain, J. Burger, J. Aberdeen, D. Connolly, and L. Hirschman. 1995. A model-theoretic coreference scoring scheme. In Proceedings of the Sixth Message understanding Conference (MUC-6), pages 45– 52, San Francisco, CA. Morgan Kaufmann Publishers. find our card
  17. X. Yang and J. Su. 2007. Coreference resolution using semantic relatedness information from automatically discovered patterns. In Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics (ACL), pages 528–535. find our card
  18. X. Yang, J. Su, G. Zhou, and C. Tan. 2004a. Improving pronoun resolution by incorporating coreferential information of candidates. In Proceedings of the 42nd Annual Meeting of the Association for Computational Linguistics (ACL), pages 127–134, Barcelona. find our card
  19. X. Yang, J. Su, G. Zhou, and C. Tan. 2004b. An NP-cluster approach to coreference resolution. In Proceedings of the 20th International Conference on Computational Linguistics, pages 219–225, Geneva. find
  20. G. Zhou and J. Su. 2000. Error-driven HMM-based chunk tagger with context-dependent lexicon. In Proceedings of the Joint Conference on Empirical Methods in Natural Language Processing and Very Large Corpora, pages 71–79, Hong Kong. find
  21. G. Zhou and J. Su. 2002. Named Entity recognition using a HMM-based chunk tagger. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics (ACL), pages 473–480, Philadelphia. find

↑ back to top

This page in Markdown →