Large Language Model for Assignment Feedback on Open-ended Subjective Questions

Le résumé fourni par la source

Feedback is a key factor in motivating and consolidating learning, but in classroom teaching, a teacher needs to provide timely and effective feedback on the homework of dozens of students, which puts much pressure on the teacher. Meanwhile, existing automatic feedback systems are not suitable for open writing tasks. The emergence of ChatGPT has attracted the attention of researchers. We selected 28 open-ended subjective question assignments to input into ChatGPT, and compared and analyzed the scores and comments generated by ChatGPT with those of teachers, to explore the feasibility of using a Large Language Model to provide timely feedback for open-ended subjective questions. Our research indicates that ChatGPT can score and evaluate learners' open-ended subjective homework, and the rating of ChatGPT can be similar to that of teachers. Moreover, ChatGPT's comments can be similar in terms of focus and emotional bias to those of teachers, indicating that ChatGPT's ratings and comments have high credibility.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.