Мақола

Methods and Algorithms of POS-tagging of Adverbs and Pronouns in Uzbek Texts

Elov Botir BoltayevichTashkent State University of Uzbek Language and Literature Named Alisher Navo’i,Dept. of Computational Linguistics and Digital Technologies,Tashkent,UzbekistanDilrabo BakhronovaUzbekistan State World Languages University,Professor, Department of Applied Sciences of Spanish Language,Tashkent,UzbekistanKhudayberganov Nizomaddin Uktambay o’g’liTashkent State University of Uzbek Language and Literature Named After Alisher Navo’i,Teacher of Computer Linguistics and Digital Technologies,Tashkent,UzbekistanSvetlana UmirovaSamarkand State University,Department of Uzbek Linguistics Samarkand State University,Samarkand,UzbekistanKarimova Zilola Dilmurod KiziTashkent State University of Uzbek Language and Literature Named After Alisher Navo’i,Master Student of Computer Linguistics,Tashkent,UzbekistanMansurova Shahinabonu Najmiddin QiziTashkent State University of Uzbek Language and Literature Named After Alisher Navo’i,Master Student of Computer Linguistics,Tashkent,Uzbekistan

2025

ABI

Аннотация

In this scientific article, the problem of automatic identification of adverbs and pronouns in Uzbek texts is analyzed based on linguistic and computational approaches. For the Uzbek language, which has an agglutinative structure, the task of POS-tagging causes not only morphological, but also syntactic difficulties. In particular, the multifunctional and contextual variability of adverbs and pronouns requires precise approaches to their automatic classification. The article analyzes performance indicators between traditional rule-based algorithms, as well as modern models based on statistical and neural networks (Conditional Random Fields and BiLSTM). Methodologically, the formal classification of models, their functionality and experimental foundations. Based on the results, the advantages and weaknesses of each approach are identified, and proposals for the optimal tagging model for the Uzbek language are put forward. The research results will be of great importance for morphological analysis, machine translation, automatic text indexation, and improving the overall quality of Uzbek NLP systems.

Ҳали таржима қилинмаган

Мавзулар

Education, Innovation and Language Studies Economic and Industrial Development

Идентификаторлар

DOI: 10.1109/ubmk67458.2025.11206966

Иқтибослар ва манбалар

0 та иқтибос3 та фойдаланилган манба

Кўрсаткичлар — AkademScholar