THE PROBLEM OF COMPUTER MODELING OF ORTHOGRAPHIC TRANSLITERATION
Keywords:
Transliteration, Orthographic Transliteration, Latin, Cyrillic, F1-score, BLEU (Bilingual Evaluation Understudy), CER (Character Error Rate) va WER (Word Error Rate)Abstract
Orthographic transliteration is the process of transliterating text based on the spelling rules of a specific natural language. The article examines the problems of transliterating text in the Cyrillic-Latin and Latin-Cyrillic directions and its computer modeling based on the rules of Uzbek spelling
References
[1] K. Knight and J. Graehl, "Machine transliteration," in Proceedings of the 35th Annual Meeting of the Association for Computational Linguistics (ACL), Madrid, July 1997, pp. 128–135. [Online]. Available: https://doi.org/10.3115/976909.979634
[2] A. Finch and E. Sumita, "Phrase-based machine transliteration," in Proceedings of the Workshop on Technologies and Corpora for Asia-Pacific Speech Translation (TCAST), 2008.
[3] B. Mansurov and A. Mansurov, "Uzbek cyrillic-latin-cyrillic machine transliteration," arXiv preprint, arXiv:2101.05162, 2021.
[4] U. Salaev, E. Kuriyozov, and C. Gómez-Rodríguez, "A machine transliteration tool between Uzbek alphabets," arXiv preprint, arXiv:2205.09578, 2022.
[5] T. Togayev, G. Tavaldiyeva, and M. Akromova, Spelling dictionary of the Uzbek language in Cyrillic and Latin alphabets. Tashkent: Editorial Office of the “Sharq” Publishing and Printing Joint-Stock Company, 1999. (In Uzbek).
