This article delves into the types of errors found in the products of machine translation in comparison with those of human translators. The focus of comparison is placed on the translation of specific contextual information that reflects the social, ...
This article delves into the types of errors found in the products of machine translation in comparison with those of human translators. The focus of comparison is placed on the translation of specific contextual information that reflects the social, political, and ideological aspects of the original text. This information should be taken into consideration by both human and machine translators engaged in Korean-English translation, especially in the case of detective novels. This article is aimed at studying the difference between human translators and machine translators when translating contextual situations. In this regard, analysis will be conducted based on the typologies of machine translation errors suggested by Seo Bo-Hyun and Kim Soonyoung. In order to discover any potential difference between human translators and machine translators, this article compares the texts translated by humans and machine translators. Currently, literary translation, such as the specific genre of detective novels, is unlikely to be subjected to automated translation that provides text translation based on computer algorithm without human involvement. However, there is a possibility that automated translation of texts like detective novels will be more common in the future if there is a huge amount of training data available including specific contextual information and highly sophisticated algorithm can be designed by computer scientists. The aim of this study is to analyze the types of errors in the machine translation of detective novels. It was revealed that the error types found in English-Korean machine translation of detective novels were limited because of the small size of the data. (Dongguk University Seoul, Korea)