본문 바로가기
  • Home

An analysis of research trends on machine translation in the linguistic and Japanese language society in Korea

  • The Japanese Language Association of Korea
  • Abbr : JLAK
  • 2022, (71), pp.125-143
  • DOI : 10.14817/jlak.2022.71.125
  • Publisher : The Japanese Language Association Of Korea
  • Research Area : Humanities > Japanese Language and Literature
  • Received : December 31, 2021
  • Accepted : March 3, 2022
  • Published : March 20, 2022

Hae-mi, Lee 1

1경희대학교

Accredited

ABSTRACT

In this study, I reviewed the current research trend of machine translation by overviewing the development of computer science and the overall previous research in linguistics. To this end, the research trend of Japanese, an individual language was identified, and the research direction was suggested by finding areas where research is insufficient or nonexistent. First, focusing on the translation method, I investigated the development process of computer engineering for the development of machine translation. Next, I analyzed the overall research trends in the humanities community in Korea since 2016. Based on its analysis results, I divided the machine translation-related researches into three phases: 'Observation' of studies that analyze errors in the machine translation, and identify limits or find potentials; 'Complement' of researches that pursue ways to improve the quality of machine translation, and use machine translation for education and research; and 'Acceptance' of researches that investigate users’ perceptions. Finally, I collected and analyzed all studies related to the Korean-Japanese and Japanese-Korean machine translation existing in Korea. The result showed that researches before 2016 mainly focused on the linguistic auxiliary analysis of morpheme, syntax, and semantic levels for RBMT or SMT. After 2016, researches have been focused on 'Acceptance' rather than the 'Observation' and 'Complement' research. The research direction appeared to be biased from a receptive point of view that accepts machine translation and actively uses it in real life without sufficient prior research that would add linguistic knowledge in order to analyze the current state of NMT which is a new form of machine translation, or solve problems for better translation outcomes. There has been no comprehensive analysis of machine translation-related research trends in the Japanese language academic community yet. It is expected that this paper will provide a research basis for researchers who take machine translation as research subjects at a stage where research has not been sufficiently accumulated.

Citation status

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