Elamur Research⌕
☰
home / papers / 2022 / The Persian Dependency Treebank Made Universal

The Persian Dependency Treebank Made Universal

Summary

Persian lacks large syntactic resources in the Universal Dependencies format. Before this work, its only public UD treebank was the news-only Uppsala treebank of about 6K sentences. The authors take the larger, multi-genre Persian Dependency Treebank (PerDT) and convert it to UD with a rule-based pipeline that includes manual handling of special cases. The pipeline has four stages:

  1. Re-tokenizing to UD conventions, which means splitting multiword verb forms and pronominal clitics.
  2. Mapping PerDT part-of-speech tags to UD tags.
  3. Making systematic revisions to the original annotation.
  4. Translating PerDT's 43 relations into UD labels using conditional rules, some of which swap heads and dependents.

Proper nouns are recovered with a BERT-based named-entity tagger and then reviewed by hand. Errors that the pipeline flags are also fixed manually. A parser trained on the converted treebank reaches 85.2 LAS. A delexicalized model trained on it transfers to English about 2 LAS points better than one trained on the Uppsala data, which suggests closer agreement with UD guidelines. Persian thus gains a bigger and more consistent resource for parsing and cross-lingual research.

Key points

  • The converted treebank has 29,107 sentences, nearly five times the 5,997 of the Uppsala Persian UD treebank (Seraji et al., 2016). It also covers several genres and a wider vocabulary.
  • In supervised parsing, the data gives a labeled attachment F-score of 85.2.
  • In delexicalized Persian-to-English parser transfer, it beats the Uppsala treebank by roughly 2% absolute LAS.
  • Systematic changes include:
    • reversing the attachment direction of verbal conjunctions;
    • retagging "million"/"billion"-type words as numbers;
    • using surface lemmas for forms of "šodan".
  • PROPN tags come from IDEN-tagged phrases plus person and location entities found by a BERT-based Persian NER tagger (Taher et al., 2020). Note that the paper's first author is Mohammad Sadegh Rasooli, with Pegah Safari as an equal-contribution co-author.

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 · Part-of-speech tagging

Cite

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

@inproceedings{safari2022persian,
  title = {{The Persian Dependency Treebank Made Universal}},
  author = {Safari, Pegah and Rasooli, Mohammad Sadegh and Moloodi, Amirsaeid and Nourian, Alireza},
  booktitle = {Proceedings of the Thirteenth Language Resources and Evaluation Conference},
  year = {2022},
  doi = {10.63317/38roncj29zvw},
  url = {https://doi.org/10.63317/38roncj29zvw}
}

References

35 references, 8 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Ammar, W., Mulcaire, G., Ballesteros, M., Dyer, C., and Smith, N. A. (2016). Many languages, one parser. Transactions of the Association for Computational Linguistics, 4:431–444. find our card
  2. Booij, G. (2012). The grammar of words: An introduction to linguistic morphology. find
  3. Oxford University Press. Chi, E. A., Hewitt, J., and Manning, C. D. (2020). Finding universal grammatical relations in multilingual bert. arXiv preprint arXiv:2005.04511. find
  4. Dadegan Research Group. (2012). Persian dependency treebank, annotation manual and user guide. Supreme Council of Information and Communication Technology (SCICT), Tehran, Iran. find
  5. Feely, W., Manshadi, M., Frederking, R., and Levin, L. (2014). The CMU METAL Farsi NLP approach. In Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC’14), pages 4052–4055, Reykjavik, Iceland, May. European Language Resources Association (ELRA). find
  6. Ghayoomi, M. and Kuhn, J. (2014). Converting an HPSG-based treebank into its parallel dependency-based treebank. In LREC, pages 802–809. find
  7. Grave, E., Bojanowski, P., Gupta, P., Joulin, A., and Mikolov, T. (2018). Learning word vectors for 157 languages. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC-2018), Miyazaki, Japan, May. European Languages Resources Association (ELRA). find our card
  8. Hewitt, J. and Manning, C. D. (2019). A structural probe for finding syntax in word representations. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 4129– 4138,… find our card
  9. Karimi-Doostan, G. (2011). Separability of light verb constructions in persian. Studia Linguistica, 65(1):70–95. find
  10. Khallash, M., Hadian, A., and Minaei-Bidgoli, B. (2013). An empirical study on the effect of morphological and lexical features in Persian dependency parsing. In Proceedings of the Fourth Workshop on Statistical Parsing of Morphologically-Rich Languages, pages 97–107, Seattle, Washington, USA, Octob… find
  11. McDonald, R., Petrov, S., and Hall, K. (2011). Multi-source transfer of delexicalized dependency parsers. In Proceedings of the conference on empirical methods in natural language processing, pages 62–72. Association for Computational Linguistics. find our card
  12. 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 Proceedings of the 51st Annual Meeting of the Assoc… find
  13. Mirzaei, A. and Moloodi, A. (2016). Persian proposition bank. In Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC 2016), pages 3828–3835, Portorož, Slovenia, May. European Language Resources Association (ELRA). find
  14. Mirzaei, A. and Safari, P. (2018). Persian discourse treebank and coreference corpus. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC-2018), Miyazaki, Japan, May. European Languages Resources Association (ELRA). find
  15. Moloodi, A. and Kouhestani, M. (2017). The role of metaphor and metonymy in the semantics of persian adjectival preverbs: A cognitive linguistics approach. Language Art, 2(2):91–105. find
  16. Naseem, T., Chen, H., Barzilay, R., and Johnson, M. (2010). Using universal linguistic knowledge to guide grammar induction. In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, pages 1234–1244, Cambridge, MA, October. Association for Computational Linguistics.… find
  17. Nourian, A., Rasooli, M. S., Imany, M., and Faili, H. (2015). On the importance of ezafe construction in Persian parsing. In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 2: Sh… find
  18. Pakzad, A. and Minaei-Bidgoli, B. (2016). An improved joint model: Pos tagging and dependency parsing. Journal of AI and Data Mining, 4(1):1–8. find
  19. Pouramini, A. and Mozayani, N. (2007). An annotation scheme for a persian treebank. Proceedings of Computational Linguistics In the Netherlands, CLIN. find
  20. Rasooli, M. S. and Tetreault, J. (2015). Yara parser: A fast and accurate dependency parser. arXiv preprint arXiv:1503.06733. — [arXiv:1503.06733](https://arxiv.org/abs/1503.06733) find
  21. Rasooli, M. S., Faili, H., and Minaei-Bidgoli, B. (2011a). Unsupervised identification of Persian compound verbs. In Mexican International Conference on Artificial Intelligence, pages 394–406. find
  22. Springer. Rasooli, M. S., Moloodi, A., Kouhestani, M., and Minaei-Bidgoli, B. (2011b). A syntactic valency lexicon for Persian verbs: The first steps towards Persian dependency treebank. In 5th Language & Technology Conference (LTC): Human Language Technologies as a Challenge for Computer Science an… find
  23. Rasooli, M. S., Kouhestani, M., and Moloodi, A. (2013). Development of a Persian syntactic dependency treebank. In Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 306–314, Atlanta, Georgia, June. As… find
  24. Seraji, M., Megyesi, B., and Nivre, J. (2012). Bootstrapping a Persian dependency treebank. Linguistic Issues in Language Technology, 7(18). find
  25. Seraji, M., Jahani, C., Megyesi, B., and Nivre, J. (2014). A Persian treebank with Stanford typed dependencies. In Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC-2014), pages 796–801, Reykjavik, Iceland, May. European Languages Resources Association (ELR… find
  26. Seraji, M., Ginter, F., and Nivre, J. (2016). Universal dependencies for Persian. In Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC 2016), pages 2361– 2365, Portorož, Slovenia, May. European Language Resources Association (ELRA). find
  27. Silveira, N., Dozat, T., de Marneffe, M.-C., Bowman, S., Connor, M., Bauer, J., and Manning, C. (2014). A gold standard dependency corpus for English. In Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC-2014), pages 2897–2904, Reykjavik, Iceland, May. Euro… find
  28. Straka, M. and Straková, J. (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, August. Association for Computational Linguistics. find our card
  29. Täckström, O., McDonald, R., and Uszkoreit, J. (2012). Cross-lingual word clusters for direct transfer of linguistic structure. In Proceedings of the 2012 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 477–487, Montréal,… find our card
  30. Täckström, O., Das, D., Petrov, S., McDonald, R., and Nivre, J. (2013). Token and type constraints for cross-lingual part-of-speech tagging. Transactions of the Association for Computational Linguistics, 1:1–12. find
  31. Taher, E., Hoseini, S. A., and Shamsfard, M. (2020). Beheshti-NER: Persian named entity recognition using BERT. arXiv preprint arXiv:2003.08875. find
  32. Tenney, I., Das, D., and Pavlick, E. (2019). BERT rediscovers the classical NLP pipeline. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 4593–4601, Florence, Italy, July. Association for Computational Linguistics. find our card
  33. Zeman, D. and Resnik, P. (2008). Cross-language parser adaptation between related languages. In Proceedings of the IJCNLP-08 Workshop on NLP for Less Privileged Languages. Zeman, D., Hajič, J., Popel, M., Potthast, M., Straka, M., Ginter, F., Nivre, J., and Petrov, S. (2018). find
  34. CoNLL 2018 shared task: Multilingual parsing from raw text to universal dependencies. In Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, pages 1–21, Brussels, Belgium, October. Association for Computational Linguistics. find our card
  35. Zhang, M., Zhang, Y., and Fu, G. (2019). Cross-lingual dependency parsing using code-mixed Tree-Bank. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 997–1006, Ho… find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/safari-2022-persian-dependency-treebank-made-universal/