본문 바로가기
  • Home

A Study on Parallel Corpus of Korean-Chinese Newspaper Headlines online: A Trial Study for Machine Translation

Hwang Eun Ha 1

1연세대학교

Accredited

ABSTRACT

Studies on translation of newspaper headlines from Korean to Chinese are first conducted. To see if parallel corpus of Korean-Chinese newspaper headlines is applicable for machine translation, Korean newspaper headlines and their Chinese correspondents are analyzed through construction and annotation of parallel corpus. The corpus covers one-month newspaper headlines of three Korean major newspapers online(Dong-a Ilbo, Chosun Ilbo and JoongAng Ilbo). The translation strategies of Korean-Chinese newspaper headlines are classified into three categories, and proportions of each category are calculated: literal translations(36.7%) and free translations(45.7%) that can be applied for machine translation, the rewrite translations that can not be applied for machine translation(17.6%). The literal translations and free translations that can be applied for machine translation(82.4%), are further analyzed four translation strategies, such as addition, omission, transposition and change of mood. Finally, aforementioned strategies are tried to implement to Korean-Chinese machine-translation.

Citation status

* References for papers published after 2023 are currently being built.