Abstract
We present lemmatization experiments on the unstandardized low-resourced languages Low Saxon and Occitan using two machine-learning-based approaches represented by MaChAmp and Stanza. We show different ways to increase training data by leveraging historical corpora, small amounts of gold data and dictionary information, and discuss the usefulness of this additional data. In the results, we find some differences in the performance of the models depending on the language. This variation is likely to be partly due to differences in the corpora we used, such as the amount of internal variation. However, we also observe common tendencies, for instance that sequential models trained only on gold-annotated data often yield the best overall performance and generalize better to unknown tokens.- Anthology ID:
- 2023.vardial-1.17
- Volume:
- Tenth Workshop on NLP for Similar Languages, Varieties and Dialects (VarDial 2023)
- Month:
- May
- Year:
- 2023
- Address:
- Dubrovnik, Croatia
- Editors:
- Yves Scherrer, Tommi Jauhiainen, Nikola Ljubešić, Preslav Nakov, Jörg Tiedemann, Marcos Zampieri
- Venue:
- VarDial
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 163–173
- Language:
- URL:
- https://aclanthology.org/2023.vardial-1.17
- DOI:
- 10.18653/v1/2023.vardial-1.17
- Cite (ACL):
- Aleksandra Miletić and Janine Siewert. 2023. Lemmatization Experiments on Two Low-Resourced Languages: Low Saxon and Occitan. In Tenth Workshop on NLP for Similar Languages, Varieties and Dialects (VarDial 2023), pages 163–173, Dubrovnik, Croatia. Association for Computational Linguistics.
- Cite (Informal):
- Lemmatization Experiments on Two Low-Resourced Languages: Low Saxon and Occitan (Miletić & Siewert, VarDial 2023)
- PDF:
- https://preview.aclanthology.org/ingest-acl-2023-videos/2023.vardial-1.17.pdf