Abstract
This paper presents a technique that employs linguistic resources to refine PoS tagging using the Universal Dependencies (UD) model. The technique is based on the development and use of lists of non-ambiguous single tokens and non-ambiguous co-occuring tokens in Portuguese (regardless of whether they constitute multiword expressions or not). These lists are meant to automatically correct the tags for such tokens after tagging. The technique is applied over the output of two well-known state of the art systems - UDPipe and UDify - and the results for a real data set have shown a significant improvement of annotation accuracy. Overall, we improve tagging accuracy by up to 1.4%, and, in terms of the number of fully correct tagged sentences, our technique produces results that are 13.9% more accurate than the corresponding original system.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The ADV class is not exactly a closed one as, similarly to English with the ending -ly, in Portuguese it is possible to turn adjectives into adverbs by adding -mente at the end (for example, the adjective final can be turned into the adverb finalmente - “finally”), but, for the purpose of our technique, we ignore such adverbs.
- 2.
References
Droganova, K., Zeman, D.: Towards deep universal dependencies. In: Proceedings of the 5th International Conference on Dependency Linguistics (Depling, SyntaxFest), pp. 144–152 (2019)
Fonseca, E.R., G Rosa, J.L., Aluísio, S.M.: Evaluating word embeddings and a revised corpus for part-of-speech tagging in Portuguese. J. Braz. Comput. Soc. 21(1), 1–14 (2015)
Kondratyuk, D., Straka, M.: 75 languages, 1 model: parsing universal dependencies universally. In: Proceedings of the Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pp. 2779–2795 (2019)
Ling, W., et al.: Finding function in form: compositional character models for open vocabulary word representation. In: Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 1520–1530 (2015)
Nivre, J., Fang, C.T.: Universal dependency evaluation. In: Proceedings of the NoDaLiDa 2017 Workshop on Universal Dependencies (UDW), pp. 86–95 (2017)
Nivre, J., et al.: Universal Dependencies v1: a multilingual treebank collection. In: Proceedings of the 10th International Conference on Language Resources and Evaluation (LREC), pp. 1659–1666 (2016)
Rademaker, A., Chalub, F., Real, L., Freitas, C., Bick, E., De Paiva, V.: Universal dependencies for Portuguese. In: Proceedings of the 4th International Conference on Dependency Linguistics (Depling), pp. 197–206 (2017)
Rehbein, I., Hirschmann, H.: POS tagset refinement for linguistic analysis and the impact on statistical parsing. In: Henrich, V., Hinrichs, E., de Kok, D., Osenova, P., Przepiorkowski, A. (eds.) Proceedings of the 13th International Workshop on Treebanks and Linguistic Theories (TLT13), pp. 172–183. University of Tübingen (2018)
Rogers, A., Kovaleva, O., Rumshisky, A.: A primer in BERTology: what we know about how BERT works. Trans. Assoc. Computat. Linguist. 8, 842–866 (2020)
Santana, M.: Kaggle - news of the Brazilian newspaper. https://www.kaggle.com/marlesson/news-of-the-site-folhauol. Accessed 14 June 2021
Straka, M.: UDPipe 2.0 prototype at CoNLL 2018 UD shared task. In: Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, pp. 197–207 (2018)
Straka, M., Straková, J.: 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, pp. 88–99 (2017)
Sulubacak, U.: Implementing universal dependency, morphology, and multiword expression annotation standards for Turkish language processing. Turk. J. Electr. Eng. Comput. Sci. 26(3), 1662–1672 (2018)
Universal Dependencies: UD Portuguese Bosque - UD version 2. https://universaldependencies.org/treebanks/pt_bosque/index.html. Accessed 14 June 2021
Acknowledgments
This work was carried out at the Center for Artificial Intelligence (C4AI-USP), with support of the São Paulo Research Foundation (FAPESP grant #2019/07665-4) and the IBM Corporation.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Lopes, L., Duran, M.S., Pardo, T.A.S. (2021). Universal Dependencies-Based PoS Tagging Refinement Through Linguistic Resources. In: Britto, A., Valdivia Delgado, K. (eds) Intelligent Systems. BRACIS 2021. Lecture Notes in Computer Science(), vol 13074. Springer, Cham. https://doi.org/10.1007/978-3-030-91699-2_41
Download citation
DOI: https://doi.org/10.1007/978-3-030-91699-2_41
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-91698-5
Online ISBN: 978-3-030-91699-2
eBook Packages: Computer ScienceComputer Science (R0)