The Impact of Translation Quality on Semantic Accuracy in Literary Texts during Computational Processing

Abstract

The purpose of this study is to analyse the influence of translation quality on the accuracy of meaning transfer in literary texts processed through computational methods. A comparative analysis method is applied to translations generated by automated systems. The findings show that machine translation often fails to preserve semantic accuracy, especially with polysemous words or complex cultural contexts. Translation systems struggle with emotional nuance and stylistic features, leading to errors in contextual interpretation and meaning distortion. Neural network-based systems offer improvements but still cannot fully convey semantic content, particularly in texts with idiomatic expressions, metaphors, and abstract language. This study provides recommendations for improving machine translation, focusing on preserving semantic accuracy and cultural nuance in literary texts.