Universal Dependencies v1: A Multilingual Treebank Collection

What it is about

The paper is the reference for the first UD release. It sets out the design principles: dependency-based and lexicalist annotation, the recoverability principle (every syntactic word must map back to the raw text), content words as the backbone of the tree with function words as dependents, and a closed inventory of 17 POS tags, the morphological features, and 40 relations. Language-specific subtypes such as acl:relcl may be added and can always be stripped to the core label. The v1.2 release (November 2015) has 37 treebanks in 33 languages, with statistics per treebank (sentences, tokens, lemmas, features, relations). The authors also describe the CoNLL-U format, the validation tools and the enhanced representation, which was then used by only one treebank (Finnish).

Ideas

• I1. en ud-eval with recoverability: check that the tokens of every output sentence rejoin exactly to the raw text, using the SpaceAfter and MISC conventions of en/src/tok.rs; measure the share of sentences that fail on EWT test, which should be zero.
• I2. en ud-eval core-label score: strip the subtype after the colon from both gold and output and report LAS beside the full LAS, to show how much of the loss comes from subtyping (acl:relcl, nmod:poss); measure the difference on EWT test.
• I3. treeagree gates: check that auxiliaries and copulas depend on a lexical predicate, which is the paper's design rule; measure how often aux or cop arcs point the other way in silver trees, per 1k sentences, with treeagree check.
• I4. en/src/udreg.rs: the paper's 17 POS tags and 40 relations are the historical base of the registry; check that the registry gate accepts every relation and feature the en crate writes, and report any that only exist in the mova dialect.

For Mova

Touches languages/en/20-syntax-overview, 10-word-classes, 12-pronouns-determiners-numerals and 13-verbs, and the en crate (udreg.rs, tok.rs, ud.rs), tools/treeagree and the mova dialect (The Mova dialect: a UD metadialect, treebanks, and converters). This is a background paper: its numbers are the UD v1 baseline, not Mova targets. Measure first the recoverability check (I1), because it is cheap and it fails loudly.

Topics: Universal Dependencies · Treebanks · Part-of-speech tagging

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.63317/2hnj7od8afy8

@inproceedings{nivre2016universal,
  title = {{Universal Dependencies v1: A Multilingual Treebank Collection}},
  author = {Nivre, Joakim and Marneffe, Marie-Catherine de and Ginter, Filip and Goldberg, Yoav and Hajič, Jan and Manning, Christopher D. and McDonald, Ryan and Petrov, Slav and Pyysalo, Sampo and Silveira, Natalia and Tsarfaty, Reut and Zeman, Daniel},
  booktitle = {Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC'16)},
  year = {2016},
  doi = {10.63317/2hnj7od8afy8},
  url = {https://doi.org/10.63317/2hnj7od8afy8}
}

References

30 references, 6 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Bosco, C., Montemagni, S., and Simi, M. (2013). Converting Italian treebanks: Towards an Italian Stanford dependency treebank. In LAW & Interoperability with Discourse. find
2. Bresnan, J. (2001). Lexical-Functional Syntax. Blackwell, Oxford. find
3. Laura Brugé, et al., editors. (2012). Functional Heads. The Cartography of Syntactic Structures, Volume 7. Oxford University Press. find
4. Buchholz, S. and Marsi, E. (2006). CoNLL-X shared task on multilingual dependency parsing. In CoNLL, pages 149–164. find our card
5. Chang, P.-C., Tseng, H., Jurafsky, D., and Manning, C. D. (2009). Discriminative reordering with Chinese grammatical relations features. In SSST, pages 51–59. find
6. Das, D. and Petrov, S. (2011). Unsupervised part-of-speech tagging with bilingual graph-based projections. In ACL, pages 600–609. find
7. de Marneffe, M.-C. and Manning, C. D. (2008). The Stanford typed dependencies representation. In Workshop on Cross-framework and Cross-domain Parser Evaluation. find our card
8. de Marneffe, M.-C., MacCartney, B., and Manning, C. D. (2006). Generating typed dependency parses from phrase structure parses. In LREC. find our card
9. de Marneffe, M.-C., Dozat, T., Silveira, N., Haverinen, K., Ginter, F., Nivre, J., and Manning, C. D. (2014). Universal Stanford dependencies: A cross-linguistic typology. In LREC, pages 4585–4592. find our card
10. Dixon, R. M. (2009). Basic Linguistic Theory. Volume 1: Methodology. Oxford University Press. ↗
11. Haverinen, K., Nyblom, J., Viljanen, T., Laippala, V., Kohonen, S., Missilä, A., Ojala, S., Salakoski, T., and Ginter, F. (2013). Building the essential resources for Finnish: the Turku dependency treebank. Language Resources and Evaluation. In press. Available online. ↗
12. Kromann, M. T. (2003). The Danish Dependency Treebank and the DTAG treebank tool. In TLT, pages 217– 220. find
13. Lipenkova, J. and Souček, M. (2014). Converting Russian dependency treebank to Stanford typed dependencies representation. In EACL. find
14. Luotolahti, J., Kanerva, J., Pyysalo, S., and Ginter, F. (2015). SETS: scalable and efficient tree search in dependency graphs. In NAACL Demo, pages 51–55. find
15. McDonald, R. and Nivre, J. (2007). Characterizing the errors of data-driven dependency parsing models. In Proceedings of EMNLP-CoNLL, pages 122–131. find our card
16. McDonald, R., Nivre, J., Quirmbach-Brundage, Y., Goldberg, Y., Das, D., Ganchev, K., Hall, K., Petrov, S., Zhang, H., Täckström, O., Bedini, C., Bertomeu Castelló, N., and Lee, J. (2013). Universal dependency annotation for multilingual parsing. In ACL. find
17. Nivre, J. and Megyesi, B. (2007). Bootstrapping a Swedish treeebank using cross-corpus harmonization and annotation projection. In TLT, pages 97–102. find
18. Osborne, T. and Maxwell, D. (2015). A historical overview of the status of function words in dependency grammar. In Proceedings of the Third International Conference on Dependency Linguistics (Depling 2015), pages 241–250. find
19. Petrov, S., Das, D., and McDonald, R. (2012). A universal part-of-speech tagset. In LREC. find
20. Pyysalo, S. and Ginter, F. (2014). Collaborative development of annotation guidelines with application to Universal Dependencies. In SLTC. find
21. Rosa, R., Mašek, J., Mareček, D., Popel, M., Zeman, D., and Žabokrtský, Z. (2014). HamleDT 2.0: Thirty dependency treebanks stanfordized. In LREC, pages 2334– 2341. find
22. Schuster, S. and Manning, C. D. (2016). Enhanced english universal dependencies: An improved representation for natural language understanding tasks. In Proceedings of the 10th International Conference on Language Resources and Evaluation. find our card
23. Seraji, M., Jahani, C., Megyesi, B., and Nivre, J. (2013). Uppsala Persian dependency treebank annotation guidelines. Technical report, Uppsala University. find
24. Stenetorp, P., Pyysalo, S., Topić, G., Ohta, T., Ananiadou, S., and Tsujii, J. (2012). Brat: a web-based tool for NLP-assisted text annotation. In ACL Demo, pages 102– 107. find
25. Štěpánek, J. and Pajas, P. (2010). Querying diverse treebanks in a uniform way. In LREC. find
26. Tesnière, L. (1959). Éléments de syntaxe structurale. Editions Klincksieck. ↗
27. Tsarfaty, R. (2013). A unified morpho-syntactic scheme of Stanford dependencies. In ACL. find
28. Zeman, D. and Resnik, P. (2008). Cross-language parser adaptation between related languages. In IJCNLP-08 Workshop on NLP for Less Privileged Languages, pages 35–42. find
29. Zeman, D., Mareček, D., Popel, M., Ramasamy, L., Štěpánek, J., Žabokrtský, Z., and Hajič, J. (2012). HamleDT: To parse or not to parse? In LREC, pages 2735– 2741. find
30. Zeman, D. (2008). Reusable tagset conversion using tagset drivers. In LREC, pages 213–218. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/ud-2016-nivre-universal-dependencies-v1-multilingual-treebank/