COMPARING HUMAN TRANSLATION WITH DEEPL TRANSLATE IN TRANSLATING IDIOM OF WEDNESDAY SERIES

This research aims to compare the strategy and accuracy of human translation and DeepL Translate in translating idioms of Wednesday series. The writers collected the idiom found in the series, analyzed the idiom’s translation strategies by Mona Baker theory, and assessed translation accuracy of human translation and DeepL Translate using Nababan et al. theory. This research employs a qualitative methodology. The writers found 38 data and the findings are as follow: the findings from human translation shows 2,6% and DeepL Translate 7,9% were using an idiom of similar meaning and form. Human translation 5,3% and DeepL Translate 2,6% using an idiom of similar meaning but dissimilar form. Translation by paraphrase strategy by human translation 92,1% and DeepL Translate 42,1%. Translation by omission of a play on idiom strategy by DeepL Translate 47,4%. The significant differences happened in the idiom translation accuracy between human translation and DeepL Translate. The assessment of accuracy level in human translation by rater 1 and rater 2 are 100%. The assessment of accuracy level in DeepL Translate are 57,9% by rater 1 and rater 2, the less accurate 5,3% by rater 1 and 10,5% by rater 2, not accurate 36,8% by rater 1 and 31,6% by rater 2.