데이터 증강이 외국어 영어 학습용 AI 챗봇의 작동 정확성과 학습자의 만족도에 미치는 영향 (Effects of data augmentation on the accuracy of response in AI chatbot for L2 English learning and the learners’ satisfaction level)

서지정보

· 발행기관 : 한국멀티미디어언어교육학회
· 수록지 정보 : 멀티미디어 언어교육 / 25권 / 2호 / 152 ~ 181페이지
· 저자명 : 김낙훈

초록

This study investigated the effects of data augmentation on the accuracy of AI chatbot responses for L2 English learning and learners’ satisfaction. Data augmentation is a computational method that can help increase the response accuracy of AI chatbot to enhance the user satisfaction. However, research on L2 learning has not been focused on evaluating the influence of data augmentation on the AI chatbot response accuracy and learner satisfaction. To clarify this influence, 36 Korean high-school English L2 learners were invited to participate in this study. The participants used two hands-on AI chatbots, with and without data augmentation, for English L2 learning. The accuracy of the AI chatbot responses was statistically analyzed, and the participants’ satisfaction in using the AI chatbot was qualitatively and quantitatively investigated. The results demonstrated that the responses of the data-augmented AI chatbot were more accurate than those of the chatbot without data augmentation. Additionally, learners who used the data-augmented AI chatbot were more satisfied with their AI chatbot-based learning experience. This research highlights the necessity of incorporating data augmentation in the design and implementation of AI chatbots for English language learning.

영어초록

This study investigated the effects of data augmentation on the accuracy of AI chatbot responses for L2 English learning and learners’ satisfaction. Data augmentation is a computational method that can help increase the response accuracy of AI chatbot to enhance the user satisfaction. However, research on L2 learning has not been focused on evaluating the influence of data augmentation on the AI chatbot response accuracy and learner satisfaction. To clarify this influence, 36 Korean high-school English L2 learners were invited to participate in this study. The participants used two hands-on AI chatbots, with and without data augmentation, for English L2 learning. The accuracy of the AI chatbot responses was statistically analyzed, and the participants’ satisfaction in using the AI chatbot was qualitatively and quantitatively investigated. The results demonstrated that the responses of the data-augmented AI chatbot were more accurate than those of the chatbot without data augmentation. Additionally, learners who used the data-augmented AI chatbot were more satisfied with their AI chatbot-based learning experience. This research highlights the necessity of incorporating data augmentation in the design and implementation of AI chatbots for English language learning.

참고자료

· 없음