Machine Translation Versus Human Translation: An Evaluation of Kurdish Machine Translation 

Bnar Abdulrazaq Azeez & Soran Mustafa Kurdi

English Language Teaching Department, Tishk International University, Kurdistan Region, Iraq

Email: [email protected]

Published: February 22, 2023

Abstract

Machine translation has made significant progress in recent years, but it still lacks the ability to fully replicate the accuracy and nuance of human translation. This is particularly true when it comes to idioms, which are expressions that have a figurative meaning that is different from the literal meaning of the words used. This study evaluates the effectiveness of Kurdish machine translation, specifically Google Translation and Bing Translation, in translating English idioms. A set of idioms, 277 to be specific, were selected and translated by Google Translation and Bing Translation. The translations were then compared to the same set of idioms translated and published by a human translator. The results of the study showed that human translation was significantly more accurate and fluent than machine translation in translating English idioms into Kurdish. While machine translation was able to provide a basic translation of the idioms, it frequently produced results that were literal and did not accurately convey the intended meaning or tone. In contrast, human translation was able to accurately convey the figurative meaning of the idioms and use appropriate Kurdish idioms as substitutes. Overall, this study highlights the limitations of machine translation in accurately conveying the meaning and tone of idiomatic expressions in language translation. 

KEYWORDS: Translation, human translation, machine translation, idiom translation, equivalency in translation. 

DOI: 10.23918/vesal2023v14