Elamur ResearchβŒ•
☰
home / papers / 2014 / Detecting Subevent Structure for Event Coreference Resolution

Detecting Subevent Structure for Event Coreference Resolution

Summary

Most work on event coreference only asks whether two event mentions denote the very same event. Araki et al. argue that this misses partial identity, especially subevent relations, where several events (e.g., capturing, killing, destroying) are parts of a larger parent event (an attack). They propose a two-stage method. First, an L2-regularized pairwise logistic regression model sorts each mention pair into one of four classes: full coreference, subevent parent-child (with direction), subevent sister, or no coreference. Second, because sister links proved fairly precise, the sisters are grouped into subevent clusters, and a voting procedure picks the most probable parent for each cluster from mentions outside it. The approach is tested on 65 news articles from the Intelligence Community domain and scored with a four-class extension of BLANC. The summed-probability voting variant raises parent-child BLANC F1 by about 3.2 points over the first-stage model. This matters because identifying subevent links can remove misleading candidates from full coreference chains, and it is presented as the first systematic treatment of this problem.

Key points

  • Corpus: 65 IC-domain newspaper articles (49 train+dev, 16 test) with 35,908 mention-pair relations: 1,253 FC, 1,198 SP, 538 SS and 32,919 NC. Fleiss' kappa is 0.620 for FC and 0.467 for SP.
  • SP extension: SP links were also derived transitively. When A is part of B and B corefers with C, A is likewise treated as a subevent of C.
  • First-stage model: pairwise multiclass logistic regression with 135 features, MetaCost to handle class imbalance, and 5-fold cross-validation.
  • First-stage scores (BLANC F1): FC 70.01, SP 56.19, SS 61.49 and NC 64.02. SS had a positive-link precision of 66.67, which motivated the second stage.
  • Second-stage parent voting:
    • Option 1 uses the single highest pairwise SP probability and reached 58.74 F1 (+2.5).
    • Option 2 sums each candidate parent's probabilities over all sisters and reached 59.45 F1 (+3.2).
    • The authors attribute remaining errors largely to abstract or rare parent expressions.

Our summary of the paper, in our own words β€” not its text. The work belongs to its authors; read it at the links above.

Fields: Linguistics & languages

Topics: Coreference resolution

Cited by

28 papers of our library cite it (of the 140673 whose bibliographies we read); 9 of them have a page here, 19 not yet.

Cite

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

@inproceedings{araki2014detecting,
  title = {{Detecting Subevent Structure for Event Coreference Resolution}},
  author = {Araki, Jun and Liu, Zhengzhong and Hovy, Eduard and Mitamura, Teruko},
  booktitle = {Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14)},
  year = {2014},
  doi = {10.63317/5m7r2rxqgcmk},
  url = {https://doi.org/10.63317/5m7r2rxqgcmk}
}

References

32 references, 12 of them with a card of ours. β†— leads to the work, Β«findΒ» searches for one we have no id for yet.

  1. Amit Bagga and Breck Baldwin. 1999. Cross-Document Event Coreference: Annotations, Experiments, and Observations. In Proceedings of ACL 1999 Workshop on Coreference and Its Applications, pages 1–8. find our card
  2. Niranjan Balasubramanian, Stephen Soderland, Mausam, and Oren Etzioni. 2013. Generating Coherent Event Schemas at Scale. In Proceedings of EMNLP 2013, pages 1721–1731. find our card
  3. Cosmin Bejan and Sanda Harabagiu. 2008. A Linguistic Resource for Discovering Event Structures and Resolving Event Coreference. In Proceedings of LREC 2008. find our card
  4. Cosmin Bejan and Sanda Harabagiu. 2010. Unsupervised Event Coreference Resolution with Rich Linguistic Features. In Proceedings of ACL 2010, pages 1412–1422. find our card
  5. Eric Bengtson and Dan Roth. 2008. Understanding the Value of Features for Coreference Resolution. In Proceedings of EMNLP 2008, pages 294–303. find our card
  6. Nathanael Chambers and Dan Jurafsky. 2008. Unsupervised Learning of Narrative Event Chains. In Proceedings of ACL-HLT 2008, pages 789–797. find
  7. Nathanael Chambers and Dan Jurafsky. 2009. Unsupervised Learning of Narrative Schemas and their Participants. In Proceedings of ACL-IJCNLP 2009, pages 602– 610. find
  8. Zheng Chen, Heng Ji, and Robert Haralick. 2009. A Pairwise Event Coreference Model, Feature Impact and Evaluation for Event Coreference Resolution. In Proceedings of RANLP 2009 Workshop on Events in Emerging Text Types (eETTs), pages 17–22. find our card
  9. Timothy Chklovski and Patrick Pantel. 2004. VerbOcean: Mining the Web for Fine-Grained Semantic Verb Relations. In Proceedings of EMNLP 2004, pages 33–40. find
  10. Ronan Collobert, Jason Weston, Léon Bottou, Michael Karlen, Koray Kavukcuoglu, and Pavel Kuksa. 2011. Natural Language Processing (Almost) from Scratch. Journal of Machine Learning Research, 12:2493–2537. find our card
  11. Agata Cybulska and Piek Vossen. 2012. Using Semantic Relations to Solve Event Coreference in Text. In In LREC 2012 Workshop on Semantic Relations-II Enhancing Resources and Applications, pages 60–67. find
  12. Dipanjan Das, Nathan Schneider, Desai Chen, and Noah A. Smith. 2010. Probabilistic Frame-Semantic Parsing. In Proceedings of NAACL-HLT 2010, pages 948–956. find our card
  13. Pedro Domingos. 1999. MetaCost: A General Method for Making Classifiers Cost-Sensitive. In Proceedings of SIGKDD 1999, pages 155–164. find
  14. Radu Florian, John F. Pitrelli, Salim Roukos, and Imed Zitouni. 2010. Improving Mention Detection Robustness to Noisy Input. In Proceedings of EMNLP 2010, pages 335–345. find
  15. Kartik Goyal, Sujay Kumar Jauhar, Huiying Li, Mrinmaya Sachan, Shashank Srivastava, and Eduard Hovy. 2013. A Structured Distributional Semantic Model for Event Co-reference. In Proceedings of ACL 2013, pages 467– 473. find our card
  16. Graeme Hirst and David St-Onge. 1998. Lexical Chains as Representations of Context for the Detection and Correction of Malapropisms. In Christiane Fellbaum, editor, WordNet: An Electronic Lexical Database, pages 305– 332, Cambridge, Massachusetts. The MIT Press. find
  17. Eduard Hovy, Teruko Mitamura, Felisa Verdejo, Jun Araki, and Andrew Philpot. 2013. Events are Not Simple: Identity, Non-Identity, and Quasi-Identity. In Proceedings of NAACL-HLT 2013 Workshop on Events: Definition, Detection, Coreference, and Representation, pages 21–28. find
  18. Kevin Humphreys, Robert Gaizauskas, and Saliha Azzam. 1997. Event Coreference for Information Extraction. In Proceedings of ACL/EACL 1997 Workshop on Operational Factors in Practical, Robust Anaphora Resolution for Unrestricted Texts, pages 75–81. find our card
  19. Silja Huttunen, Roman Yangarber, and Ralph Grishman. 2002. Complexity of Event Structure in IE Scenarios. In Proceedings of COLING 2002, pages 1–7. find
  20. Jay J. Jiang and David W. Conrath. 1997. Semantic Similarity Based on Corpus Statistics and Lexical Taxonomy. In Proceedings of International Conference Research on Computational Linguistics (ROCLING X), pages 19–33. find
  21. Claudia Leacock and Martin Chodorow. 1998. Combining Local Context and WordNet Similarity for Word Sense Identification. In Christiane Fellfaum, editor, WordNet: An Electronic Lexical Database, pages 265–283, Cambridge, Massachusetts. The MIT Press. find
  22. 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 CoNLL-2011 Shared Task, pages 28–34. find our card
  23. Michael Lesk. 1986. Automatic Sense Disambiguation Using Machine Readable Dictionaries: How to Tell a Pine Cone from an Ice Cream Cone. In Proceedings of SIGDOC 1986, pages 24–26. find
  24. Dekang Lin. 1998. An Information-Theoretic Definition of Similarity. In Proceedings of ICML 1998, pages 296– 304. find
  25. Pablo Mendes, Max Jakob, and Christian Bizer. 2012. DBpedia: A Multilingual Cross-domain Knowledge Base. In Proceedings of LREC 2012. find
  26. Sameer S. Pradhan, Lance Ramshaw, Ralph Weischedel, Jessica MacBride, and Linnea Micciulla. 2007. Unrestricted Coreference: Identifying Entities and Events in OntoNotes. In Proceedings of the International Conference on Semantic Computing (ICSC), pages 446–453. find
  27. Marta Recasens and Eduard Hovy. 2011. BLANC: Implementing the Rand index for coreference evaluation. Natural Language Engineering, 17(4):485–510. find
  28. Michaela Regneri, Alexander Koller, and Manfred Pinkal. 2010. Learning Script Knowledge with Web Experiments. In Proceedings of ACL 2010, pages 979–988. find
  29. Philip Resnik. 1995. Using Information Content to Evaluate Semantic Similarity in a Taxonomy. In Proceedings of IJCAI 1995, pages 448–453. find our card
  30. Roger C. Schank and Robert P. Abelson. 1977. Scripts, Plans, Goals, and Understanding: An Inquiry into Human Knowledge Structures. Lawrence Erlbaum Associates. find
  31. Stephen Tratz and Eduard Hovy. 2011. A Fast, Accurate, Non-Projective, Semantically-Enriched Parser. In Proceedings of EMNLP 2011, pages 1257–1268. find
  32. Zhibiao Wu and Martha Palmer. 1994. Verbs Semantics and Lexical Selection. In Proceedings of ACL 1994, pages 133–138. find

↑ back to top

This page in Markdown β†’