맞춤형 GPT를 활용한 연구형 수업 적용 연구 한외대조언어학세미나 수업을 중심으로

Abstract

With the rapid advancement of artificial intelligence(AI), GPT (Generative Pre-trained Transformer) has emerged as a transformative tool in the field of education, offering capabilities such as text generation, information summarization, and interactive feedback. This study examines the application of a customized GPT model in the Seminar on Contrastive Linguistics between Korean and Foreign Languages, a doctoral-level course offered at Busan University of Foreign Studies. The customized GPT served as a learning assistant and feedback tool, supporting students in preparing presentations, conducting pre-class research, and participating in discussions. The study analyzed student interactions with GPT over nine weeks, focusing on the types of prompts students used, including grammatical structures, translation and semantic interpretation, error analysis, and educational strategies. A survey was also conducted to assess GPT's usability in terms of learning support, ease of use, interactivity, content quality, and overall satisfaction. The findings revealed that GPT was highly effective in enhancing students' understanding of key concepts (90% satisfaction), improving usability with an intuitive interface and quick response times (80% satisfaction), and providing relevant and practical learning strategies (70% satisfaction). However, areas requiring improvement were also identified, including the need for increased response clarity, better citation of information sources, and enhanced support for complex and ambiguous queries. This study demonstrates the potential of GPT as an effective learning assistant in higher education, particularly in contrastive linguistics, and highlights key areas for future refinement to maximize its educational value.