COMPUTATIONAL APPROACHES AND LINGUISTIC ANNOTATION STRATEGIES FOR CONSTRUCTING UZBEK–CHINESE PARALLEL CORPUS RESOURCES: DATA COLLECTION, SENTENCE ALIGNMENT, AND QUALITY EVALUATION

Authors

  • Xiaoqi YOU National University of Uzbekistan (PhD)

Keywords:

parallel corpus, Uzbek, Chinese, sentence alignment, sentence embeddings, corpus annotation, quality evaluation, low-resource machine translation

Abstract

Uzbek–Chinese is a low-resource language pair for which large, curated parallel corpora remain scarce, despite growing academic, economic and cultural exchange between Uzbekistan and China. This article examines computational approaches and linguistic annotation strategies for constructing Uzbek–Chinese parallel corpus resources. It reviews data collection channels, preprocessing problems specific to the two languages, embedding-based sentence alignment, filtering methods and annotation layers, and it proposes a quality evaluation framework that combines manual auditing with automatic and downstream measures. The analysis draws on recent work on multilingual sentence embeddings, linear-time sentence alignment, audits of web-crawled multilingual data and multilingual evaluation benchmarks. The article argues that the quality of an Uzbek–Chinese corpus depends less on its size than on script normalization, segmentation consistency, alignment confidence and documented human evaluation. A practical pipeline is outlined, together with annotation guidelines and indicators that allow corpus quality to be reported transparently.

References

NLLB Team, Costa-jussà M.R. et al. No Language Left Behind: Scaling Human-Centered Machine Translation. – arXiv preprint, 2022. – arXiv:2207.04672. – URL: https://arxiv.org/abs/2207.04672 (accessed: 01.11.2026).

Goyal N. et al. The Flores-101 Evaluation Benchmark for Low-Resource and Multilingual Machine Translation // Transactions of the Association for Computational Linguistics. – 2022. – Vol. 10. – P. 522–538. – DOI: 10.1162/tacl_a_00474 (accessed: 01.11.2026).

Kreutzer J. et al. Quality at a Glance: An Audit of Web-Crawled Multilingual Datasets // Transactions of the Association for Computational Linguistics. – 2022. – Vol. 10. – P. 50–72. – DOI: 10.1162/tacl_a_00447 (accessed: 01.11.2026).

Tiedemann J., Thottingal S. OPUS-MT – Building open translation services for the World // Proceedings of the 22nd Annual Conference of the European Association for Machine Translation. – Lisboa, 2020. – P. 479–480. – URL: https://aclanthology.org/2020.eamt-1.61 (accessed: 01.11.2026).

Thompson B., Koehn P. Vecalign: Improved Sentence Alignment in Linear Time and Space // Proceedings of EMNLP-IJCNLP 2019. – Hong Kong, 2019. – P. 1342–1348. – DOI: 10.18653/v1/D19-1136 (accessed: 01.11.2026).

Feng F., Yang Y., Cer D., Arivazhagan N., Wang W. Language-agnostic BERT Sentence Embedding // Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). – Dublin, 2022. – P. 878–891. – DOI: 10.18653/v1/2022.acl-long.62 (accessed: 01.11.2026).

[Author(s) not verified.] Obtaining Parallel Sentences in Low-Resource Language Pairs with Minimal Supervision. – Hindawi, 2022. – URL: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9365574/ (accessed: 01.11.2026).

[Author(s) not verified.] Filling the Gap for Uzbek: Creating Translation Resources for Southern Uzbek // Proceedings of the Tenth Conference on Machine Translation (WMT 2025). – 2025. – URL: https://aclanthology.org/2025.wmt-1.83/ (accessed: 01.11.2026).

Downloads

Published

2026-11-01

How to Cite

Xiaoqi YOU. (2026). COMPUTATIONAL APPROACHES AND LINGUISTIC ANNOTATION STRATEGIES FOR CONSTRUCTING UZBEK–CHINESE PARALLEL CORPUS RESOURCES: DATA COLLECTION, SENTENCE ALIGNMENT, AND QUALITY EVALUATION. IQRO , 24(01), 1364–1370. Retrieved from https://www.wordlyknowledge.uz/index.php/iqro/article/view/9705