English

BBPOS: BERT-based Part-of-Speech Tagging for Uzbek

Computation and Language 2025-01-20 v1 Artificial Intelligence

Abstract

This paper advances NLP research for the low-resource Uzbek language by evaluating two previously untested monolingual Uzbek BERT models on the part-of-speech (POS) tagging task and introducing the first publicly available UPOS-tagged benchmark dataset for Uzbek. Our fine-tuned models achieve 91% average accuracy, outperforming the baseline multi-lingual BERT as well as the rule-based tagger. Notably, these models capture intermediate POS changes through affixes and demonstrate context sensitivity, unlike existing rule-based taggers.

Keywords

Cite

@article{arxiv.2501.10107,
  title  = {BBPOS: BERT-based Part-of-Speech Tagging for Uzbek},
  author = {Latofat Bobojonova and Arofat Akhundjanova and Phil Ostheimer and Sophie Fellenz},
  journal= {arXiv preprint arXiv:2501.10107},
  year   = {2025}
}
R2 v1 2026-06-28T21:09:11.941Z