Elamur Research⌕
☰
home / papers / 2018 / Marrying Universal Dependencies and Universal Morphology

Marrying Universal Dependencies and Universal Morphology

Summary

UD and UniMorph each define a cross-lingual scheme for morphosyntactic annotation. UD annotates tokens in treebanks, while UniMorph lists inflected word types from Wiktionary paradigms. In theory the two could complement and check each other. In practice they diverge, because of differing design philosophies (UD is built bottom-up, UniMorph top-down), annotator choices, and conversion errors in individual datasets. The authors therefore treat the task as translating concrete resources, not just abstract schemata. They build a deterministic, rule-based converter from UD v2 features to UniMorph tags and add language-specific post-editing rules for the 31 languages covered by both projects. Checking the converted tags against UniMorph tables gives a macro-averaged recall of 64.13%, an improvement over a naive feature-to-feature mapping. Morphological tagging on UD sentences performs about the same with either tag set. The work matters because it links type- and token-level morphological data, and the interpretable rules expose errors and inconsistencies in both projects that maintainers can fix.

Key points

  • The released tool (github.com/unimorph/ud-compatibility) swaps the UD features in a CoNLL-U file for UniMorph MSDs. It needs no training and is fully deterministic.
  • Converted tags were evaluated by lookup in UniMorph for 31 languages, reaching 64.13% macro-average recall. Some languages remain poor, such as Dutch, where UD annotations lack attributes that UniMorph records.
  • The extrinsic tagging test covered eight languages: Bulgarian, Danish, Finnish, Latvian, Portuguese, Russian, Spanish and Swedish. In none of them did the UniMorph-converted tags yield a lower F1 than the original UD tags. The largest gap was in Spanish (78.31 vs 96.44); most differences were small.
  • Scheme sizes: UniMorph has at least 212 values over 23 attributes, while UD has POS plus 23 attributes and language-specific ones. UD v2.1 contains 102 treebanks in 60 languages.
  • The paper also critically lists missing attributes and cross-lingual inconsistencies in both resources. Examples include the Spanish form "vas" wrongly tagged as feminine plural, and gender annotated for "refrescante" in Portuguese but not in Spanish.

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: Universal Dependencies · Treebanks

Cited by

34 papers of our library cite it (of the 140673 whose bibliographies we read); 4 of them have a page here, 30 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/w18-6011 · arXiv 1810.06743

@inproceedings{mccarthy2018marrying,
  title = {{Marrying Universal Dependencies and Universal Morphology}},
  author = {McCarthy, Arya D. and Silfverberg, Miikka and Cotterell, Ryan and Hulden, Mans and Yarowsky, David},
  booktitle = {Proceedings of the Second Workshop on Universal Dependencies (UDW 2018)},
  year = {2018},
  doi = {10.18653/v1/w18-6011},
  eprint = {1810.06743},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.18653/v1/w18-6011}
}

References

38 references, 9 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Mark Aronoff. 1976. Word formation in generative grammar. Linguistic Inquiry Monographs Cambridge, Mass., 1:1–134. find
  2. Mark Aronoff. 2007. In the beginning was the word. Language, 83(4):803–830. find
  3. Matthew Baerman, Dunstan Brown, Greville G Corbett, et al. 2005. The syntax-morphology interface: A study of syncretism, volume 109. Cambridge University Press. find
  4. Yonatan Belinkov, Nadir Durrani, Fahim Dalvi, Hassan Sajjad, and James Glass. 2017. What do neural machine translation models learn about morphology? In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), volume 1, pages 861–872. find our card
  5. Jean Berko. 1958. The child’s learning of English morphology. Word, 14(2-3):150–177. find
  6. Joan Bresnan and Sam A. Mchombo. 1995. The lexical integrity principle: Evidence from Bantu. Natural Language & Linguistic Theory, 13(2):181–254. find
  7. Peter F Brown, Vincent J Della Pietra, Stephen A Della Pietra, and Robert L Mercer. 1993. The mathematics of statistical machine translation: Parameter estimation. Computational linguistics, 19(2):263–311. find
  8. Noam Chomsky. 1970. Remarks on nominalization. In R. Jacobs and P. Rosenbaum, editors, Reading in English Transformational Grammar, pages 184– 221. Ginn and Co., Waltham. find
  9. Ryan Cotterell and Georg Heigold. 2017. Cross-lingual character-level neural morphological tagging. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 748–759. Association for Computational Linguistics. find
  10. Ryan Cotterell, Christo Kirov, John Sylak-Glassman, Géraldine Walther, Ekaterina Vylomova, Arya D. McCarthy, Katharina Kann, Sebastian Mielke, Garrett Nicolai, Miikka Silfverberg, David Yarowsky, Jason Eisner, and Mans Hulden. 2018. The CoNLL– SIGMORPHON 2018 shared task: Universal morphological re… find
  11. Ryan Cotterell, Christo Kirov, John Sylak-Glassman, Gėraldine Walther, Ekaterina Vylomova, Patrick Xia, Manaal Faruqui, Sandra Kübler, David Yarowsky, Jason Eisner, et al. 2017. find
  12. CoNLL-SIGMORPHON 2017 shared task: Universal morphological reinflection in 52 languages. Proceedings of the CoNLL–SIGMORPHON 2017 Shared Task: Universal Morphological Reinflection, pages 1–30. find
  13. Ryan Cotterell, Christo Kirov, John Sylak-Glassman, David Yarowsky, Jason Eisner, and Mans Hulden. 2016. find
  14. The SIGMORPHON 2016 shared task— morphological reinflection. In Proceedings of the 14th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology, pages 10–22. find
  15. Peter Dirix, Liesbeth Augustinus, Daniel van Niekerk, and Frank Van Eynde. 2017. Universal dependencies for Afrikaans. In Proceedings of the NoDaLiDa 2017 Workshop on Universal Dependencies, 22 May, Gothenburg Sweden, 135, pages 38–47. Linköping University Electronic Press. find
  16. Elliott Franco Drábek and David Yarowsky. 2005. Induction of fine-grained part-of-speech taggers via classifier combination and crosslingual projection. In Proceedings of the ACL Workshop on Building and Using Parallel Texts, pages 49–56. Association for Computational Linguistics. find
  17. Martin Haspelmath. 2010. Comparative concepts and descriptive categories in crosslinguistic studies. Language, 86(3):663–687. find
  18. Gholamhossein Karimi-Doostan. 2011. Separability of light verb constructions in Persian. Studia Linguistica, 65(1):70–95. find
  19. Aleksandr E. Kibrik. 1998. Archi (Caucasian– Daghestanian). Wiley Online Library. find
  20. Christo Kirov, Ryan Cotterell, John Sylak-Glassman, Géraldine Walther, Ekaterina Vylomova, Patrick Xia, Manaal Faruqui, Arya D. McCarthy, Sandra Kübler, David Yarowsky, Jason Eisner, and Mans Hulden. 2018. UniMorph 2.0: Universal Morphology. In Proceedings of the Eleventh International Conference… find
  21. Christo Kirov, John Sylak-Glassman, Rebecca Knowles, Ryan Cotterell, and Matt Post. 2017. A rich morphological tagger for English: Exploring the cross-linguistic tradeoff between morphology and syntax. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational… find our card
  22. Christo Kirov, John Sylak-Glassman, Roger Que, and David Yarowsky. 2016. Very-large scale parsing and normalization of Wiktionary morphological paradigms. In LREC. find our card
  23. Ralph B. Long. 1957. Paradigms for English verbs. Publications of the Modern Language Association of America, pages 359–372. find
  24. Chaitanya Malaviya, Matthew R. Gormley, and Graham Neubig. 2018. Neural factor graph models for cross-lingual morphological tagging. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2652–2662. Association for Computational Ling… find
  25. Mitchell P. Marcus, Mary Ann Marcinkiewicz, and Beatrice Santorini. 1993. Building a large annotated corpus of English: The Penn Treebank. Computational linguistics, 19(2):313–330. find our card
  26. Guillaume Wisniewski and Ophélie Lacroix. 2017. A systematic comparison of syntactic representations of dependency parsing. In Proceedings of the NoDaLiDa 2017 Workshop on Universal Dependencies (UDW 2017), pages 146–152. find
  27. Joakim Nivre, Željko Agić, Lars Ahrenberg, Lene Antonsen, Maria Jesus Aranzabe, Masayuki Asahara, Luma Ateyah, Mohammed Attia, Aitziber Atutxa, Liesbeth Augustinus, et al. 2017. Universal dependencies 2.1. LINDAT/CLARIN digital library at the Institute of Formal and Applied Linguistics (ÚFAL), Fa… find
  28. Slav Petrov, Dipanjan Das, and Ryan McDonald. 2012. A universal part-of-speech tagset. In Proceedings of the Eight International Conference on Language Resources and Evaluation (LREC’12), Istanbul, Turkey. European Language Resources Association (ELRA). find
  29. Martin Popel, Zdenĕk Žabokrtský, and Martin Vojtek. 2017. Udapi: Universal API for universal dependencies. In Proceedings of the NoDaLiDa 2017 Workshop on Universal Dependencies (UDW 2017), pages 96–101. find our card
  30. Ivan A. Sag, Timothy Baldwin, Francis Bond, Ann Copestake, and Dan Flickinger. 2002. Multiword expressions: A pain in the neck for NLP. In International Conference on Intelligent Text Processing and Computational Linguistics, pages 1–15. Springer. find our card
  31. Andrew Spencer. 1991. Morphological theory: An introduction to word structure in generative grammar, volume 2. Basil Blackwell Oxford. find
  32. Milan Straka and Jana Straková. 2017. Tokenizing, POS tagging, lemmatizing and parsing UD 2.0 with UDPipe. In Proceedings of the CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, pages 88–99, Vancouver, Canada. Association for Computational Linguistics. find our card
  33. John Sylak-Glassman. 2016. The composition and use of the universal morphological feature schema (Uni-Morph schema). Technical report, Department of Computer Science, Johns Hopkins University. find
  34. John Sylak-Glassman, Christo Kirov, David Yarowsky, and Roger Que. 2015. A language-independent feature schema for inflectional morphology. In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Proces… find
  35. Oscar Täckström, Dipanjan Das, Slav Petrov, Ryan Mc-Donald, and Joakim Nivre. 2013. Token and type Daniel Zeman. 2008. Reusable tagset conversion using tagset drivers. In LREC, volume 2008, pages 28–30. find our card
  36. Daniel Zeman, Ondřej Dušek, David Mareček, Martin Popel, Loganathan Ramasamy, Jan Štěpánek, Zdeněk Žabokrtskỳ, and Jan Hajič. 2014. Hamledt: Harmonized multi-language dependency treebank. Language Resources and Evaluation, 48(4):601– 637. find
  37. Daniel Zeman, Martin Popel, Milan Straka, Jan Hajic, Joakim Nivre, Filip Ginter, Juhani Luotolahti, Sampo Pyysalo, Slav Petrov, Martin Potthast, et al. 2017. find
  38. CoNLL 2017 shared task: multilingual parsing from raw text to universal dependencies. Proceedings of the CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, pages 1–19. find our card

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/mccarthy-2018-marrying-universal-dependencies-universal-morphology/