Description
This study investigates the differences between human translation and AI-assisted translation through a corpus-based error analysis of translations produced by English-major students, Google Translate, and ChatGPT. Drawing on a learner-oriented translation corpus, the study aims to identify and classify common translation errors, compare their frequency and severity across translation types, and evaluate the strengths and limitations of human and machine-generated translations. Error categories are analyzed using an established translation quality framework, with particular attention to lexical choice, grammatical accuracy, mistranslation, omission, and contextual appropriateness. The findings are expected to reveal distinct error patterns associated with each translation approach and provide insights into the complementary roles of human translators and AI tools. The study also discusses pedagogical implications for translation training and post-editing practices in English language education, contributing to the growing field of human–machine collaboration in translation.
Từ khóa
corpus-based translation studies; machine translation; human translation; Google Translate; ChatGPT; error analysis; post-editing; human–AI collaboration; translation pedagogy.
Thông tin các tác giả
Trần Trung Hiếu: TS, đang công tác tại Trường Đại học Ngoại ngữ - Tin học TP. HCM, 828 Sư Vạn Hạnh, Quận 10, TP. HCM, email: hieutt1@huflit.edu.vn
