본문 바로가기
  • Home

A Study on Generative AI-Based Feedback Techniques for Tutoring Beginners’ Error Codes on Online Judge Platforms

  • Journal of The Korea Society of Computer and Information
  • Abbr : JKSCI
  • 2024, 29(8), pp.191-200
  • Publisher : The Korean Society Of Computer And Information
  • Research Area : Engineering > Computer Science
  • Received : June 3, 2024
  • Accepted : July 26, 2024
  • Published : August 30, 2024

Juyeon Lee 1 Seung-Hyun Kim 1

1한국교원대학교

Accredited

ABSTRACT

The rapid advancement of computer technology and artificial intelligence has significantly impacted software education in Korea. Consequently, the 2022 revised curriculum demands personalized education. However, implementing personalized education in schools is challenging. This study aims to facilitate personalized education by utilizing incorrect codes and error information submitted by beginners to construct prompts. And the difference in the frequency of correct feedback generated by the generative AI model and the prompts was examined. The results indicated that providing appropriate error information in the prompts yields better performance than relying solely on the excellence of the generative AI model itself. Through this research, we hope to establish a foundation for the realization of personalized education in programming education in Korea.

Citation status

* References for papers published after 2023 are currently being built.

This paper was written with support from the National Research Foundation of Korea.