Elamur Research⌕
☰
home / papers / 2015 / Knowledge-lean projection of coreference chains across languages

Knowledge-lean projection of coreference chains across languages

Summary

Building coreference resolvers usually requires hand-crafted, language-specific rules or large annotated datasets, and most low-resourced languages have neither. Grishina and Stede test how well coreference annotations can be transferred from English into another language using only automatic alignment, with no linguistic knowledge of the target language. They built a small parallel corpus in English, German and Russian. It covers newswire (Project Syndicate) and short stories, plus medicine leaflets from EMEA/OPUS for English-German only. All texts were annotated under shared, largely language-neutral guidelines modelled on OntoNotes and PoCoS. Sentences were aligned with HunAlign, and words with GIZA++ trained on moderately sized newswire data. Mentions and chains were then projected directly. Projection worked better for English-Russian than for English-German, and stories scored higher than newswire. The error analysis links many failures to structural mismatches, such as relative clauses rendered as participles, possessors turning into adjectives, or English pronouns translated as German articles. The study provides a baseline showing that cheap projection is feasible, and it identifies where limited syntactic knowledge could improve results.

Key points

  • The corpus contains 38 parallel texts in three genres, annotated in the MMAX-2 tool. Inter-annotator agreement for English and German was κ ≈ 0.86–0.87 (binary overlap), with MUC of 77.28 (English) and 73.91 (German). Medical leaflets showed lower agreement (κ = 0.76, MUC 70%).
  • GIZA++ was trained on about 200,000 English-German and 170,000 English-Russian sentence pairs. On English-German gold data, the intersective alignment reached P 92.95, R 51.23 and F 66.05, somewhat below Padó (2007) and Spreyer (2011), which used much larger Europarl training sets.
  • With full mention spans, average MUC F1 for the projected chains was 54.6 in German and 71.0 in Russian. Precision was 68.0/82.1 and recall 45.8/62.6.
  • When only RE heads (minimal spans) were evaluated, average F1 rose to 74.1 for German and 81.3 for Russian. This compares with 80.5 reported by Postolache et al. (2006) for English-Romanian, a setup that used a dedicated aligner and POS/syntax rules.
  • Newswire scored lowest, which the authors attribute to more complex NPs. Medical leaflets suffered from borderline referential cases, such as diseases and body parts.

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); 6 of them have a page here, 10 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/w15-3403

@inproceedings{grishina2015knowledge,
  title = {{Knowledge-lean projection of coreference chains across languages}},
  author = {Grishina, Yulia and Stede, Manfred},
  booktitle = {Proceedings of the Eighth Workshop on Building and Using Comparable Corpora},
  year = {2015},
  doi = {10.18653/v1/w15-3403},
  url = {https://doi.org/10.18653/v1/w15-3403}
}

References

31 references, 10 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Amit Bagga and Breck Baldwin. 1998. Algorithms for scoring coreference chains. In The first international conference on language resources and evaluation workshop on linguistics coreference, volume 1, pages 563–566. find our card
  2. František Čermák and Alexandr Rosen. 2012. The case of InterCorp, a multilingual parallel corpus. International Journal of Corpus Linguistics, 17(3):411–427. find
  3. Maud Ehrmann, Marco Turchi, and Ralf Steinberger. 2011. Building a multilingual named entity-annotated corpus using annotation projection. In RANLP, pages 118–124. find
  4. Markus Gärtner, Anders Björkelund, Gregor Thiele, Wolfgang Seeker, and Jonas Kuhn. 2014. Visualization, search, and error analysis for coreference annotations. In Proceedings of 52nd Annual Meeting of the Association for Computational Linguistics: System Demonstrations, pages 7–12. find
  5. Liane Guillou, Christian Hardmeier, Aaron Smith, Jörg Tiedemann, and Bonnie Webber. 2014. ParCor 1.0: A parallel pronoun-coreference corpus to support statistical MT. In Ninth International Conference on Language Resources and Evaluation (LREC’14), pages 3191–3198. European Language Resources Associ… find our card
  6. Sanda M. Harabagiu and Steven J. Maiorano. 2000. Multilingual coreference resolution. In Proceedings of the sixth conference on Applied natural language processing, pages 142–149. Association for Computational Linguistics. find our card
  7. Eduard Hovy, Mitchell Marcus, Martha Palmer, Lance Ramshaw, and Ralph Weischedel. 2006. OntoNotes: the 90% solution. In Proceedings of the human language technology conference of the NAACL, Companion Volume: Short Papers, pages 57–60. Association for Computational Linguistics. find
  8. Rebecca Hwa, Philip Resnik, Amy Weinberg, Clara Cabezas, and Okan Kolak. 2005. Bootstrapping parsers via syntactic projection across parallel texts. Natural language engineering, 11(03):311–325. find
  9. Olga Krasavina and Christian Chiarcos. 2007. PoCoS: Potsdam coreference scheme. In Proceedings of the Linguistic Annotation Workshop, pages 156–163. Association for Computational Linguistics. find our card
  10. Kerstin Anna Kunz. 2010. Variation in English and German Nominal Coreference: A Study of Political Essays, volume 21. Peter Lang. find
  11. Xiaoqiang Luo. 2005. On coreference resolution performance metrics. In Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing, pages 25–32. Association for Computational Linguistics. find our card
  12. Rada Mihalcea, Carmen Banea, and Janyce M. Wiebe. 2007. Learning multilingual subjective language via cross-lingual projections. In Proceedings of ACL-2007. find
  13. Ruslan Mitkov and Catalina Barbu. 2002. Using bilingual corpora to improve pronoun resolution. Languages in contrast, 4(2):201–211. find
  14. Arne Neumann. 2015. discoursegraphs: A graph-based merging tool and converter for multilayer annotated corpora. In Nordic Conference of Computational Linguistics NODALIDA 2015, page 309. find
  15. Franz Josef Och and Hermann Ney. 2003. A systematic comparison of various statistical alignment models. Computational linguistics, 29(1):19–51. find
  16. Sylwia Ozdowska. 2006. Projecting POS tags and syntactic dependencies from English and French to Polish in aligned corpora. In Proceedings of the International Workshop on Cross-Language Knowledge Induction, pages 53–60. Association for Computational Linguistics. find
  17. Sebastian Padó and Mirella Lapata. 2005. Cross-linguistic projection of role-semantic information. In Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing, pages 859–866. Association for Computational Linguistics. find
  18. Sebastian Padó. 2007. Cross-lingual annotation projection models for role-semantic information. Ph.D. thesis, German Research Center for Artificial Intelligence and Saarland University. find
  19. Oana Postolache, Dan Cristea, and Constantin Orasan. 2006. Transferring coreference chains through word alignment. In Proceedings of LREC-2006. find our card
  20. Sameer Pradhan, Lance Ramshaw, Mitchell Marcus, Martha Palmer, Ralph Weischedel, and Nianwen Xue. 2011. Conll-2011 shared task: Modeling unrestricted coreference in OntoNotes. In Proceedings of the Fifteenth Conference on Computational Natural Language Learning: Shared Task, pages 1–27. Association… find our card
  21. Sameer Pradhan, Alessandro Moschitti, Nianwen Xue, Olga Uryupina, and Yuchen Zhang. 2012. CoNLL-2012 shared task: Modeling multilingual unrestricted coreference in OntoNotes. In Joint Conference on EMNLP and CoNLL-Shared Task, pages 1– 40. Association for Computational Linguistics. find our card
  22. Altaf Rahman and Vincent Ng. 2012. Translation-based projection for multilingual coreference resolution. In Proceedings of the 2012 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 720–730. Association for Computational Lin… find our card
  23. Asad Sayeed, Tamer Elsayed, Nikesh Garera, David Alexander, Tan Xu, Douglas W. Oard, David Yarowsky, and Christine Piatko. 2009. Arabic cross-document coreference detection. In Proceedings of the ACL-IJCNLP 2009 Conference Short Papers, pages 357–360. Association for Computational Linguistics. find
  24. Kathrin Spreyer. 2011. Does it have to be trees?: Data-driven dependency parsing with incomplete and noisy training data. Ph.D. thesis, Universitätsbibliothek Potsdam. find
  25. Jörg Tiedemann. 2009. News from OPUS-a collection of multilingual parallel corpora with tools and interfaces. In Recent Advances in Natural Language Processing, volume 5, pages 237–248. find
  26. Jörg Tiedemann. 2014. Rediscovering annotation projection for cross-lingual parser induction. In Proc. COLING. find
  27. Dan Tufis, Radu Ion, Alexandru Ceausu, and Dan Stefanescu. 2006. Improved lexical alignment by combining multiple reified alignments. In EACL. find
  28. Dániel Varga, Péter Halácsy, András Kornai, Viktor Nagy, László Németh, and Viktor Trón. 2007. Parallel corpora for medium density languages. Amsterdam Studies in the Theory and History of Linguistics Science series 4, 292:247. find
  29. Marc Vilain, John Burger, John Aberdeen, Dennis Connolly, and Lynette Hirschman. 1995. A model-theoretic coreference scoring scheme. In Proceedings of the 6th conference on Message understanding, pages 45–52. Association for Computational Linguistics. find our card
  30. David Yarowsky, Grace Ngai, and Richard Wicentowski. 2001. Inducing multilingual text analysis tools via robust projection across aligned corpora. In Proceedings of the first international conference on Human language technology research, pages 1– 8. Association for Computational Linguistics. find
  31. Imed Zitouni and Radu Florian. 2008. Mention detection crossing the language barrier. In Proceedings of the Conference on Empirical Methods in Natural Language Processing, pages 600–609. Association for Computational Linguistics. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/grishina-2015-knowledge-lean-projection-coreference-chains/