Elamur Research⌕
☰
home / papers / 2013 / Semantic Relations between Events and their Time, Locations and Participants for Event Coreference Resolution

Semantic Relations between Events and their Time, Locations and Participants for Event Coreference Resolution

Summary

The paper asks how much each part of an event description helps decide whether two mentions refer to the same real-world event. The authors model an event as four slots: action, participants, time and location. They argue that matching actions alone cannot settle coreference, and that partial matches matter. Such matches arise through hyponymy, part-of or membership links and shifts in granularity. Machine learning is deliberately avoided so that the effect of each factor stays transparent.

Working on the ECB corpus with KYOTO tools and WordNet, the system first builds high-recall chains of candidate actions using Leacock & Chodorow similarity. It then prunes these chains by checking whether the participants, times or locations of the candidates are compatible. Participant filtering contributes the most. Its scores come close to a strong same-lemma baseline but remain about 1% lower in F. The authors attribute the baseline's strength to how the corpus was built. They call for datasets that cover many distinct instances of the same event type.

Key points

  • Data: the EventCorefBank (ECB) corpus with the stand-off annotation of Lee et al. (2012). It holds 482 Google News texts in 43 topics, annotated for cross-document event coreference.
  • Pipeline: the KYOTO tools and the Stanford Parser handle syntax, followed by WSD against WordNet 3.0. An event ontology attached to 266 hypernyms covers 53,964 synsets. Kybot rules extracted 54,236 participant, 5,728 location and 3,435 time mentions.
  • The best setup, action plus participant similarity, reached 70.7% MUC, 74.1% B3, 64.9% CEAFm, 70.4% BLANC F and 69.8 CoNLL F1.
  • Compared with the lemma baseline, this setup gains up to 2 points of recall but loses about 4% precision. Participant similarity beats action-only similarity by about 1–6% F.
  • Time and location filtering added little. The authors partly explain this by how rarely these slots occur compared with participants.

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

10 papers of our library cite it (of the 140673 whose bibliographies we read); 3 of them have a page here, 7 not yet.

Cite

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

@inproceedings{cybulska2013semantic,
  title = {{Semantic Relations between Events and their Time, Locations and Participants for Event Coreference Resolution}},
  author = {Cybulska, Agata and Vossen, Piek},
  booktitle = {Proceedings of the International Conference Recent Advances in Natural Language Processing RANLP 2013},
  year = {2013},
  url = {https://aclanthology.org/R13-1021/}
}

References

23 references, 12 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. ACE-Event. 2005. ACE English Annotation Guidelines for Events, ver. 5.4.3 2005.07.01. find
  2. Bagga, Amit and Breck Baldwin, “Algorithms for Scoring Coreference Chains”, in Proceedings of LREC 1998 find our card
  3. Bejan, Cosmin Adrian and Sanda Harabagiu, “A Linguistic Resource for Discovering Event Structures and Resolving Event Coreference”, in Proceedings of the 6th International Conference on Language Resources and Evaluation (LREC), Marrakech, Morocco, 2008 find our card
  4. Bejan, Cosmin Adrian and Sanda Harabagiu, “Unsupervised Event Coreference Resolution with Rich Linguistic Features”, in Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics, Uppsala, Sweden, 2010 find our card
  5. Chalupsky, Hans et al., “The RACR Machine Reading System”, in prep. 2013 find
  6. Chen, Zheng and Heng Ji, “Event Coreference Resolution: Algorithm, Feature Impact and Evaluation”, in Proceedings of Events in Emerging Text Types (eETTs) Workshop, in conjunction with RANLP, Bulgaria, 2009 find
  7. Chen, Bin, Su, Jian, Pan, Sinno Jialin and Chew Lim Tan, “A Unified Event Coreference Resolution by Integrating Multiple Resolvers”, in Proceedings of the 5th International Joint Conference on Natural Language Processing, Chiang Mai, Thailand, November, 2011 find our card
  8. Cybulska, Agata and Piek Vossen, “Event models for Historical Perspectives: Determining Relations between High and Low Level Events in Text, Based on the Classification of Time, Location and Participants”, in Proceedings of LREC 2010, Valletta, Malta, May 17-23, 2010 find
  9. Cybulska, Agata and Piek Vossen, “Historical Event Extraction from Text”, in Proceedings of ACL LaTeCH, Portland, US, June 2011 find
  10. Cybulska, Agata, and Piek Vossen, “Using Semantic Relations to Solve Event Coreference in Text”, in Proceedings of the Workshop: Semantic Relations-II. Enhancing Resources and Applications (SemRel2012), Istanbul, Turkey, May 2012 van Hage, Willem Robert, Malaisé, Véronique, Segers, Roxane, Hollink,… find
  11. Harabagiu, Sanda M., Bunescu, Razvan C. and Steven J. Maiorano, “Text and Knowledge Mining for Coreference Resolution”, in Proceedings of the second meeting of the North American Chapter of the Association for Computational Linguistics on Language technologies, 2001 find
  12. Howald, Blake Stephen and Martha Abramson, “The Use of Granularity in Rhetorical prediction”, in Proceedings of the First Joint Conference on Lexical and Computational Semantics (*SEM), pages 44-48, Montreal, Canada, 2012 find
  13. Hovy, Eduard, Mitamura, Teruko, Verdejo, Felisa and Philpot, Andrew, “Identity and Quasi-Identity Relations for Event Coreference”, in prep.2013 find
  14. Ide, Nancy and David Woolner, 2007, “Historical Ontologies”, in: Ahmad, Khurshid, Brewster, Christopher, and Mark Stevenson (eds.), Words and Intelligence II: Essays in Honor of Yorick Wilks, Springer, 137-152. find
  15. Kuebler, Sandra and Denislava Zhekova, “Singletons and Coreference Resolution Evaluation”, in Proceedings of Recent Advances in NLP, Hissar, Bulgaria, September, 2011 find our card
  16. Leacock, Claudia and Martin Chodorow, “Combining local context with WordNet similarity for word sense identification”, in Christiane Fellbaum (ed.), WordNet: A lexical Reference System and its Application, MIT Press, Cambridge, MA. Lee, Heeyoung, Recasens, Marta, Chang, Angel, Surdeanu, Mihai and Da… find our card
  17. Luo, Xiaoqiang, "On coreference resolution performance metrics", in Proceedings of the Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing (EMNLP-2005), 2005 find our card
  18. Mulkar-Mehta, Rutu, Hobbs, Jerry R. and Eduard Hovy,“Granularity in Natural Language Discourse”,in Proceedings of International Conference on Computational Semantics, 2011a Mulkar-Mehta, Rutu, Hobbs, Jerry R. and Eduard Hovy, “Applications and Discovery of Granularity Structures in Natural Language… find our card
  19. Ng, Vincent and Claire Cardie, “Improving Machine Learning Approaches to Coreference Resolution”, in Proceedings of the 40 th Annual Meeting of the Association for Computational Linguistics (ACL), Philadelphia, 2002 find our card
  20. Ponzetto, Simone Paolo and Michael Strube, “Exploiting Semantic Role Labeling, WordNet and Wikipedia for Coreference Resolution”, in Proceedings of the Human Language Technology Conference of the North American Chapter of the ACL, pages 192-199, New York, 2006 find our card
  21. Pradhan, Sameer, Ramshaw, Lance, Marcus, Mitchell, Palmer, Martha, Weischedel, Ralph and Nianwen Xue, “CoNLL-2011 shared task: Modeling unrestricted coreference in OntoNotes”, in Proceedings of CoNLL 2011: Shared Task, 2011. find our card
  22. Quine, Willard V., “Events and Reification” in E. LePore and B.P. McLaughlin (eds.), Action and Events, Basil Blackwell, New York, 1985 find
  23. Recasens, Marta, Hovy, Eduard and M. Antònia Martí, “Identity, non-identity, and near-identity: Addressing the complexity of coreference”, Lingua, 121(6):1138-1152, 2011a Recasens, Marta and Eduard Hovy, "BLANC: Implementing the Rand index for coreference evaluation", Natural Language Engineering, 1… find our card

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/cybulska-2013-semantic-relations-between-events-their/