본문 바로가기
  • Home

Analysis of Passives in English-to-Korean Machine Translation Systems’ Output

  • Modern English Education
  • Abbr : MEESO
  • 2023, 24(), pp.1-9
  • DOI : 10.18095/meeso.2023.24.1.1
  • Publisher : The Modern English Education Society
  • Research Area : Humanities > English Language and Literature > English Language Teaching
  • Received : December 28, 2022
  • Accepted : January 31, 2023
  • Published : February 28, 2023

Kim, Chae-Eun 1

1부산대학교

Accredited

ABSTRACT

Artificial intelligence (AI)-based tools for teaching and learning English as a foreign language have become increasingly functional, leading to the growing use of AI translators and chatbots in fields of English education and research. This study investigated the potential of English-to-Korean machine translators (MTs) as media of interactive language learning to support Korean EFL learners’ reading comprehension development. Using translation data from research articles on English passives involving NP movement, this study evaluated how well two popular MTs could translate English passive sentences into Korean. This study analyzed MTs’ output in two ways: (i) whether the product’s meaning was accurate, and (ii) what type of Korean passive the product was. First, it calculated rates of semantically accurate and grammatically appropriate translations, both actives and passives produced by the two MTs. Second, it calculated rates of three different types of Korean passives produced by MTs. Results demonstrated that currently available English-to-Korean MTs could produce semantically correct translations at high rates in the form of Korean passives. Active voice constructions also showed significant rates. These findings suggest that MTs have strong potential as tools for English reading pedagogy.

Citation status

* References for papers published after 2022 are currently being built.