MasakhaPOS: Part-of-Speech Tagging for Typologically Diverse African Languages

Simple item page
dc.contributor.authorDione, Cheikh M. Bamba
dc.contributor.authorNabende, Peter
dc.contributor.authorMukiibi, Jonathan
dc.contributor.authorChinedu Uchechukwu
dc.contributor.authorUchechukwu, Chinedu
dc.contributor.authorAbdullahi, Muhammad
dc.contributor.authorKlakow, Dietrich
dc.date.accessioned2025-03-07T18:56:42Z
dc.date.available2025-03-07T18:56:42Z
dc.date.issued2023-05-23
dc.description.abstractIn this paper, we present MasakhaPOS, the largest part-of-speech (POS) dataset for 20 typologically diverse African languages. We discuss the challenges in annotating POS for these languages using the UD (universal dependencies) guidelines. We conducted extensive POS baseline experiments using conditional random field and several multilingual pretrained language models. We applied various cross-lingual transfer models trained with data available in UD. Evaluating on the Masakha- POS dataset, we show that choosing the best transfer language(s) in both single-source and multi-source setups greatly improves the POS tagging performance of the target languages, in particular when combined with cross-lingual parameter-efficient fine-tuning methods. Crucially, transferring knowledge from a language that matches the language family and morphosyntactic properties seems more effective for POS tagging in unseen languages.
dc.identifier.citationDione, C. M. B., Adelani, D., Nabende, P., Alabi, J., Sindane, T., Buzaaba, H., ... & Klakow, D. (2023). Masakhapos: Part-of-speech tagging for typologically diverse african languages. arXiv preprint arXiv:2305.13989.
dc.identifier.otherhttps://doi.org/10.48550/arXiv.2305.13989
dc.identifier.urihttps://nru.uncst.go.ug/handle/123456789/10073
dc.language.isoen
dc.publisherarXiv preprint arXiv
dc.titleMasakhaPOS: Part-of-Speech Tagging for Typologically Diverse African Languages
dc.typeArticle
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
2305.13989v1.pdf
Size:
458.53 KB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
Humanities and the Arts