Key points are not available for this paper at this time.
With the recent rise of large language models, learning assistance systems that utilize natural language processing capabilities are gaining attention. However, existing learning assistance systems mainly aim to provide answers to learners' questions, and have limitations in accurately identifying individuallearning needs and analyzing questions to provide customized answers. Therefore, in this study, we compare chatGPT and mentor responses in the process of learning a programming language and find out the limitations of each and develop a system to assist programming language learning using chatGPT. The developed system allows learners to get realtime answers from chatGPT immediately, and can also provide additional personalized guidance and answers based on the mentor's judgment. In addition, through the analysis of learners' responses and chatGPT's answers, we aim to evaluate the appropriateness and accuracy of the answers and explore and suggest technical limitations and improvement measures of chatGPT. By analyzing and discussing the differences between the traditional mentoring method and the learning assist method using chatGPT, we aim to diagnose the limitations of chatGPT in programming language learning. Through this study, we expect to develop an effective programming language learning assistance system by integrating the advantages of chatGPT and mentoring to improve learners' understanding and learning efficiency.
Moon et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: