A COMPARATIVE AND COMPUTATIONAL STUDY OF PROVERBS AND IDIOMS IN ENGLISH AND UZBEK
Keywords:
proverbs, idioms, translation, English, Uzbek, computational linguistics, phraseology, machine translationAbstract
Proverbs and idioms are integral parts of language and culture, encapsulating collective wisdom, values, and figurative expression. Their translation between languages poses a unique challenge due to structural, semantic, and cultural differences. This study investigates the comparative features of English and Uzbek proverbs and idioms, while also testing the capacity of machine translation (MT) systems to recognize and translate them accurately. Using a bilingual dataset of 300 idioms and proverbs, the research evaluates human translation strategies alongside the performance of neural MT systems such as Google Translate and Yandex Translate. Findings reveal that while human translators employ functional equivalence and cultural substitution to maintain idiomatic meaning, MT systems frequently produce literal translations, resulting in semantic distortions and cultural loss. The study highlights the importance of integrating computational phraseology and cultural annotation into NLP models to improve idiom recognition and translation.
References
Baker, M. (1992). In Other Words: A Coursebook on Translation. Routledge.
Fernando, C. (1996). Idioms and Idiomaticity. Oxford University Press.
Koehn, P. (2020). Neural Machine Translation. Cambridge University Press.
Mieder, W. (2004). Proverbs: A Handbook. Greenwood Press.
Sag, I. A., Baldwin, T., Bond, F., Copestake, A., & Flickinger, D. (2002). Multiword expressions: A pain in the neck for NLP. Proceedings of CICLing 2002, 1–15.
Venuti, L. (1995). The Translator’s Invisibility: A History of Translation. Routledge.
