본문 바로가기
  • Home

A Study on the Chinese Translation of Korean Polysemous Verbs

  • Journal of Chinese Language and Literature
  • 2022, (89), pp.313-332
  • DOI : 10.15792/clsyn..89.202204.313
  • Publisher : Chinese Literary Society Of Yeong Nam
  • Research Area : Humanities > Chinese Language and Literature
  • Received : March 10, 2022
  • Accepted : April 13, 2022
  • Published : April 30, 2022

Yeonok Hong 1 Liu, Yafei 2

1세종대학교
2서울대학교

Accredited

ABSTRACT

Learning polysemy is one of the difficult parts for learners studying foreign languages. This is because the meaning of polysemy with more than one expressed meaning is deeply related to the culture of the society and the cognitive method of the members of the society influenced by the culture. In the case of translation, the meaning may vary depending on the accurate translation of polysemy, or the exact transmission may not be possible. Thus, learners must also invest a lot of effort into learning polysemy when learning a foreign language. Therefore, this paper examines whether Korean Chinese learners can correctly translate sentences using a Korean polysemous verb, and the ability to process polysemous verb translation of neural network machine translation when they translate Korean polysemous verbs into Chinese. Thus, this paper is a follow-up study of the Korean-Chinese machine translation pattern of a homonymous verb conducted as a previous study. By examining the machine translation pattern of a polysemous verb, it can be confirmed that Google still shows low accuracy while Papago and Baidu have improved translation accuracy in the translation of polysemy. In addition, Papago and Baidu have much better translation skills than middle-class Chinese-level college students who participated in this study in the translation of a polysemous verb and a homonymous verb. Thus, students will be able to use the neural network translation system as useful learning tools to help modify the translation on their own in the future.

Citation status

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