Published November 1, 2026 | Version v1

THEORETICAL AND METHODOLOGICAL FOUNDATIONS OF DEVELOPING UZBEK–CHINESE PARALLEL CORPORA: LINGUISTIC ALIGNMENT, TRANSLATION EQUIVALENCE, AND CROSS-LINGUISTIC STRUCTURAL DIFFERENCES

  • 1. National University of Uzbekistan, PhD

Description

The article examines the theoretical and methodological foundations of building Uzbek–Chinese parallel corpora. Although parallel corpora have become a core resource for translation studies, contrastive linguistics and machine translation, the Uzbek–Chinese language pair remains poorly resourced. The study analyzes three interrelated problems: the alignment of texts at the document, paragraph, sentence and word levels; the identification of translation equivalence; and the structural differences between an agglutinative language with subject–object–verb order and an isolating language with subject–verb–object order. Drawing on recent work on Uzbek corpus development and on embedding-based word alignment, the article proposes a multi-level framework for corpus construction that combines automatic alignment with linguistically informed annotation and human verification. The framework is intended to support contrastive research, translator training and the development of language technologies for the Uzbek–Chinese pair.

Files

1244-1249.pdf

Files (215.4 kB)

Name Size
md5:bf7ada78ffdb4498acb6df2b21af7a93
215.4 kB Preview Download

Additional details

References

  • Amirkulov B. Uzbeki-English parallel corpus algorithm and alignment problem // Central Asian Journal of Literature, Philosophy and Culture. – 2023. – URL: https://bibsonomy.org/bibtex/2ddd0c49052c9654c8c51bc33865a91a4/centralasian_20 (accessed: 01.11.2026).
  • Dou Z.-Y., Neubig G. Word alignment by fine-tuning embeddings on parallel corpora // Proceedings of EACL 2021. – 2021. – URL: https://arxiv.org/abs/2101.08231 (accessed: 01.11.2026).
  • Jalili Sabet M., Dufter P., Yvon F., Schütze H. SimAlign: High quality word alignments without parallel training data using static and contextualized embeddings // Findings of EMNLP 2020. – 2020. – URL: https://arxiv.org/abs/2004.08728 (accessed: 01.11.2026).
  • Mamasaidov M., Aral A., Shopulatov A., Inomjonov M. Filling the gap for Uzbek: Creating translation resources for Southern Uzbek [Preprint]. – 2025. – URL: https://arxiv.org/abs/2508.14586 (accessed: 01.11.2026).
  • Sharipov M., Mattiev J., Sobirov J., Baltayev R. Creating a morphological and syntactic tagged corpus for the Uzbek language [Preprint]. – 2022. – URL: https://arxiv.org/abs/2210.15234 (accessed: 01.11.2026).