Does Universal Dependencies need a parsing representation? An investigation of English

Summary

UD makes content words the heads and attaches function words (auxiliaries, copulas, subordinators, adpositions) to them. The authors ask whether English parsing would benefit from an intermediate representation where function words are heads, with a conversion back to UD afterwards. They define reversible tree transformations for four constructions (prepositional phrases, verb groups, copular clauses and marked subordinate clauses) that are lossless on gold data, and train MaltParser on each transformed version of the UD English Web Treebank. Some variants are indeed easier to parse when scored in their own representation. After converting parser output back to UD, however, accuracy always drops: one error in a function-head tree can become two in UD, and moving dependents propagates wrong attachments. The paper is a useful negative result: parsing into a "friendlier" representation and converting does not improve English UD parsing.

How Mova uses it

• Function words are leaves of the tree: rule en.errors.aux-cop-leaf in the grammar expert system (en/src/expert.rs) — function words (aux, cop, case, mark, …) should be leaves with almost no dependents. The paper is cited as the rationale for keeping content heads even though function heads are easier for a parser.
• Indirectly, it supports the choice to train the English parser (en/src/parse.rs) directly on UD trees rather than on a converted intermediate scheme.

Effectiveness in Mova

Not measured separately. Background reading behind one annotation-checking rule and a design choice (parse UD directly); no experiment with function-head intermediate representations was run.

👨‍🔬💥

Topics: Universal Dependencies

Cite

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

@inproceedings{silveira2015does,
  title = {{Does Universal Dependencies need a parsing representation? An investigation of English}},
  author = {Silveira, Natalia and Manning, Christopher},
  booktitle = {Proceedings of the Third International Conference on Dependency Linguistics (Depling 2015)},
  year = {2015},
  url = {https://aclanthology.org/W15-2134/}
}

References

16 references, 5 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Miguel Ballesteros and Joakim Nivre. 2012. Malt-Optimizer: An Optimization Tool for MaltParser. In Proceedings of the Demonstrations at the 13th Conference of the European Chapter of the Association for Computational Linguistics, EACL ’12, pages 58–62, Stroudsburg, PA, USA. Association for Computational Linguistics. find
2. Kepa Bengoetxea and Koldo Gojenola. 2009. Exploring Treebank Transformations in Dependency Parsing. In Proceedings of the International Conference RANLP-2009, pages 33–38, Borovets, Bulgaria. Association for Computational Linguistics. find
3. Daniel M. Bikel. 2004. Intricacies of collins’ parsing model. Comput. Linguist., 30(4):479–511, December. find
4. Angelina Ivanova, Stephan Oepen, and Lilja vrelid. 2013. Survey on parsing three dependency representations for English. 51st Annual Meeting of the Association for Computational Linguistics Proceedings of the Student Research Workshop, pages 31– 37. find
5. Richard Johansson and Pierre Nugues. 2007. Extended constituent-to-dependency conversion for English. In Joakim Nivre, Heiki-Jaan Kalep, Kadri Muischnek, and Mare Koit, editors, NODALIDA 2007 Proceedings, pages 105–112, Tartu, Estonia. University of Tartu. find our card
6. Dan Klein and Christopher D. Manning. 2003. Accurate Unlexicalized Parsing. In Proceedings of the 41st Annual Meeting on Association for Computational Linguistics - Volume 1, ACL ’03, pages 423– 430, Stroudsburg, PA, USA. Association for Computational Linguistics. find our card
7. Marie-Catherine Marneffe and Christopher D. Manning. 2008. The Stanford Typed Dependencies Representation. Coling 2008: Proceedings of the workshop on Cross-Framework and Cross-Domain Parser Evaluation, pages 1–8. find our card
8. Marie-Catherine de Marneffe, Timothy Dozat, Natalia Silveira, Katri Haverinen, Filip Ginter, Joakim Nivre, and Christopher D. Manning. 2014. Universal Stanford dependencies: A cross-linguistic typology. Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC-2014). find our card
9. Ryan McDonald, Kevin Lerman, and Fernando Pereira. 2006. Multilingual dependency analysis with a two-stage discriminative parser. In Proceedings of the Tenth Conference on Computational Natural Language Learning, CoNLL-X ’06, pages 216–220, Stroudsburg, PA, USA. Association for Computational Linguistics. find
10. Jens Nilsson, Joakim Nivre, and Johan Hall. 2006. Graph Transformations in Data-driven Dependency Parsing. In Proceedings of the 21st International Conference on Computational Linguistics and the 44th Annual Meeting of the Association for Computational Linguistics, ACL-44, pages 257–264, Stroudsburg, PA, USA. Association for Computational Linguistics. find
11. Jens Nilsson, Joakim Nivre, and Johan Hall. 2007. Generalizing Tree Transformations for Inductive Dependency Parsing. Proceedings of the 45th Annual Meeting of the Association of Computational Linguistics, pages 968–975. find
12. Joakim Nivre, Johan Hall, Jens Nilsson, Atanas Chanev, Glsen Eryigit, Sandra Kbler, Svetoslav Marinov, and Erwin Marsi. 2007. MaltParser: A language-independent system for data-driven dependency parsing. Natural Language Engineering, 13(02):95–135. find
13. Joakim Nivre, Cristina Bosco, Jinho Choi, Marie-Catherine de Marneffe, Timothy Dozat, Richárd Farkas, Jennifer Foster, Filip Ginter, Yoav Goldberg, Jan Hajič, Jenna Kanerva, Veronika Laippala, Alessandro Lenci, Teresa Lynn, Christopher Manning, Ryan McDonald, Anna Missilä, Simonetta Montemagni, Slav Petrov, Sampo Pyysalo, Natalia Silveira, Maria Simi, Aaron Smith, Reut Tsarfaty, Veronika Vincze, and Daniel Zeman. 2015. Universal dependencies 1.0. find
14. Slav Petrov, Leon Barrett, Romain Thibaux, and Dan Klein. 2006. Learning Accurate, Compact, and Interpretable Tree Annotation. In Proceedings of the 21st International Conference on Computational Linguistics and the 44th Annual Meeting of the Association for Computational Linguistics, ACL-44, pages 433–440, Stroudsburg, PA, USA. Association for Computational Linguistics. find our card
15. Roy Schwartz, Omri Abend, and Ari Rappoport. 2012. Learnability-based syntactic annotation design. In COLING, volume 24, pages 2405–2422. find
16. Natalia Silveira, Timothy Dozat, Marie-Catherine de Marneffe, Samuel Bowman, Miriam Connor, John Bauer, and Christopher D. Manning. 2014. A Gold Standard Dependency Corpus for English. In Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC-2014). find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/silveira-2015-does-universal-dependencies-need-parsing/