본문 바로가기
  • Home

Designing a ChatGPT-based social mentor chatbot for adults with autism spectrum disorder: A prompt engineering approach

  • Journal of The Korea Society of Computer and Information
  • Abbr : JKSCI
  • 2025, 30(8), pp.87~102
  • Publisher : The Korean Society Of Computer And Information
  • Research Area : Engineering > Computer Science
  • Received : July 14, 2025
  • Accepted : August 4, 2025
  • Published : August 29, 2025

Seung-Yeon Kim 1 Yeseul Park 2 Bowon Lee 2 Youngsun Lee 1

1이화여자대학교
2인하대학교

Accredited

ABSTRACT

As digital social interactions become more prevalent, adults with Autism Spectrum Disorder (ASD) often struggle with interpreting implicit meanings in text-based messages conversations. This study employed a Design-Based Research (DBR) methodology to develop and refine a ChatGPT-based social mentor chatbot that supports digital communication for adults with ASD. The development proceeded through two iterative design cycles, focusing on prompt engineering to enhance the chatbot’s contextual sensitivity and social appropriateness. Usability feedback from both cycles informed the final design, resulting in a more usable and applicable tool for real-world messenger-based interactions.

Citation status

* References for papers published after 2024 are currently being built.

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