Elamur Research⌕
☰
home / papers / 2018 / Rule- and Learning-based Methods for Bridging Resolution in the ARRAU Corpus

Rule- and Learning-based Methods for Bridging Resolution in the ARRAU Corpus

Summary

This paper describes the author's two submissions to track 2 of the CRAC 2018 shared task on bridging anaphora resolution. The task is to link anaphoric noun phrases to related but non-identical antecedents, and the data is the second release of the ARRAU corpus. The author first re-implemented the rule-based system of Hou et al. (2014), which was the best-performing approach on ISNotes. On ARRAU it performed badly.

Analysis traced the drop to annotation differences. Hou's rules mostly find referential bridging, where the anaphor depends on context to be understood. ARRAU, by contrast, largely annotates lexical bridging, such as subset, element and meronymy relations between concepts.

The rules were then revised to cover lexical bridging and compared with a learning-based model that uses equivalent information. The adapted rule-based system gives reasonable results on anaphor detection, anaphora resolution and full bridging resolution in all three ARRAU domains. It beats the learner in every setting, reaching 39.8% accuracy on the main RST anaphora resolution task.

The work shows that "bridging" is not a single, consistently annotated phenomenon. This matters for anyone reusing systems or datasets across corpora.

Key points

  • ARRAU (second release): about 350K tokens and 5512 bridging anaphors across three domains:
    • RST (news): 3777
    • TRAINS (dialogue): 710
    • PEAR stories (narratives): 333
  • Data preparation: the corpus was converted from MMAX into a CoNLL-12-style format. Problematic cases were removed, including multiple or empty antecedents and discontinuous markables. This left 3303 anaphors in RST, 558 in TRAINS and 303 in PEAR.
  • Re-implementation check: the Hou et al. (2014) re-implementation scores F1 17.8 for full bridging resolution on ISNotes, close to the 18.4 originally reported.
  • Learning-based model: Weka's SMO classifier with markable and pair features, including WordNet relations, semantic connectivity and word2vec cosine similarity.
  • Comparison: the rule-based system outperformed the learner in all settings, with significance tested by a Wilcoxon signed rank test at the 0.01 level. It often scored higher on the small PEAR and TRAINS sets than on RST.

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); 7 of them have a page here, 3 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/w18-0703

@inproceedings{roesiger2018rule,
  title = {{Rule- and Learning-based Methods for Bridging Resolution in the ARRAU Corpus}},
  author = {Roesiger, Ina},
  booktitle = {Proceedings of the First Workshop on Computational Models of Reference, Anaphora and Coreference},
  year = {2018},
  doi = {10.18653/v1/w18-0703},
  url = {https://doi.org/10.18653/v1/w18-0703}
}

References

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

  1. Emmanuel Lassalle and Pascal Denis. 2011. Leveraging different meronym discovery methods for bridging resolution in french. Anaphora Processing and Applications, pages 35–46. find
  2. Nicholas Asher and Alex Lascarides. 1998. Bridging. Journal of Semantics, 15(1):83–113. find
  3. Stefan Baumann and Arndt Riester. 2012. Referential and Lexical Givenness: semantic, prosodic and cognitive aspects. In Gorka Elordieta and Pilar Prieto, editors, Prosody and Meaning, number 25 in Interface Explorations. Mouton de Gruyter, Berlin. ↗
  4. Katja Markert, Yufang Hou, and Michael Strube. 2012. Collective classification for fine-grained information status. In Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics: Long Papers-Volume 1, pages 795–804, Jeju Island, Korea. Association for Computational Linguistics. find our card
  5. Aoife Cahill and Arndt Riester. 2012. Automatically acquiring fine-grained information status distinctions in German. In Proceedings of the 13th Annual Meeting of the Special Interest Group on Discourse and Dialogue, pages 232–236. Association for Computational Linguistics. find
  6. Tomas Mikolov, Kai Chen, Greg Corrado, and Jeffrey Dean. 2013. Efficient estimation of word representations in vector space. CoRR, abs/1301.3781. ↗
  7. Anna Nedoluzhko, Jiří Mírovskỳ, Radek Ocelák, and Jiří Pergler. 2009. Extended coreferential relations and bridging anaphora in the prague dependency treebank. In Proceedings of the 7th Discourse Anaphora and Anaphor Resolution Colloquium (DAARC 2009), Goa, India, pages 1–16. find
  8. Herbert H. Clark. 1975. Bridging. In Proceedings of the 1975 workshop on Theoretical issues in natural language processing, pages 169–174. Association for Computational Linguistics. find
  9. Malvina Nissim, Shipra Dingare, Jean Carletta, and Mark Steedman. 2004. An annotation scheme for information status in dialogue. Proceedings of The fourth international conference on Language Resources and Evaluation. find
  10. Kari Fraurud. 1990. Definiteness and the processing of noun phrases in natural discourse. Journal of Semantics, 7(4):395–433. ↗
  11. Yulia Grishina. 2016. Experiments on bridging across languages and genres. In Proceedings of the first Workshop on Coreference Resolution Beyond OntoNotes (NAACL-HLT), pages 7–15, San Diego, USA. find our card
  12. Massimo Poesio. 2004. Discourse annotation and semantic annotation in the gnome corpus. In Proceedings of the 2004 ACL Workshop on Discourse Annotation, DiscAnnotation ’04, pages 72–79, Stroudsburg, PA, USA. Association for Computational Linguistics. find
  13. John A Hawkins. 1978. Definiteness and indefiniteness: A study in reference and grammaticality prediction. atlantic highlands. find
  14. Massimo Poesio and Ron Artstein. 2008. Anaphoric Annotation in the ARRAU Corpus. In International Conference on Language Resources and Evaluation (LREC), Marrakech, Morocco. find our card
  15. Jerry R Hobbs, Mark E Stickel, Douglas E Appelt, and Paul Martin. 1993. Interpretation as abduction. Artificial Intelligence, 63:69–142. find
  16. Massimo Poesio, Rahul Mehta, Axel Maroudas, and Janet Hitzeman. 2004. Learning to resolve bridging references. In Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics, page 143. Association for Computational Linguistics. find our card
  17. Yufang Hou. 2016. Unrestricted Bridging Resolution. Ph.D. thesis. find our card
  18. Yufang Hou, Katja Markert, and Michael Strube. 2013a. Cascading collective classification for bridging anaphora recognition using a rich linguistic feature set. In Proceedings of the Conference on Empirical Methods in Natural Language Processing, pages 814–820, Seattle, USA. find our card
  19. Massimo Poesio and Renata Vieira. 1998. A corpus-based investigation of definite description use. Computational Linguistics, 24(2):183–216. ↗
  20. Massimo Poesio, Renata Vieira, and Simone Teufel. 1997. Resolving bridging references in unrestricted text. In Proceedings of a Workshop on Operational Factors in Practical, Robust Anaphora Resolution for Unrestricted Texts, pages 1–6. Association for Computational Linguistics. find our card
  21. Yufang Hou, Katja Markert, and Michael Strube. 2013b. Global inference for bridging anaphora resolution. In Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 907–917, Atlanta, USA. find our card
  22. Altaf Rahman and Vincent Ng. 2012. Learning the fine-grained information status of discourse entities. In Proceedings of the 13th Conference of the European Chapter of the Association for Computational Linguistics, EACL ’12, pages 798–807, Stroudsburg, PA, USA. Association for Computational Linguistics. find
  23. Yufang Hou, Katja Markert, and Michael Strube. 2014. A rule-based system for unrestricted bridging resolution: Recognizing bridging anaphora and finding links to antecedents. In Proceedings of the Conference on Empirical Methods in Natural Language Processing, pages 2082–2093, Seattle, USA. find our card
  24. Arndt Riester and Stefan Baumann. 2017. The RefLex Scheme – Annotation guidelines. SinSpeC. Working papers of the SFB 732 Vol. 14, University of Stuttgart. ↗
  25. Anna Roitberg and Anna Nedoluzhko. 2016. Bridging corpus for russian in comparison with czech. In CORBON@ HLT-NAACL, pages 59–66. find our card
  26. Sidney Siegel and N. John Jr. Castellan. 1988. Non-parametric Statistics for the Behavioral Sciences, 2nd edition. McGraw-Hill, Berkeley, CA. find
  27. 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. Olga Uryupina, Ron Artstein, Antonella Bristot, Fered-erica Cavicchio, Francesca Delogu, Kepa Rodriguez, and Massimo Poesio. to appear. Annotating a broad range of anaphoric phenomena, in a variety of genres: the arrau corpus. Journal of Natural Language Engineering. ↗ our card
  28. Renata Vieira and Simone Teufel. 1997. Towards resolution of bridging descriptions. In Proceedings of the eighth conference on European chapter of the Association for Computational Linguistics, pages 522–524. Association for Computational Linguistics. find our card
  29. Šárka Zikánová, Eva Hajicová, Barbora Hladká, Pavlína Jínová, Jirí Mírovskỳ, Anja Nedoluzhko, Lucie Poláková, Katerina Rysová, Magdaléna Rysová, and Jan Václ. 2015. Discourse and coherence. From the Sentence Structure to Relations in Text. Institute of Formal and Applied Linguistics. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/roesiger-2018-rule-learning-based-methods-bridging/