Prompt engineering to improve the performance of teaching and learning materials Recommendation of Generative Artificial Intelligence

In this study, prompt engineering that improves prompts was explored to improve the performance of teaching and learning materials recommendations using generative artificial intelligence such as GPT and Stable Diffusion. Picture materials were used as the types of teaching and learning materials. T...

Full description

Saved in:
Bibliographic Details
Published in韓國컴퓨터情報學會論文誌 Vol. 28; no. 8; pp. 195 - 204
Main Authors Soo-Hwan Lee, Ki-Sang Song
Format Journal Article
LanguageKorean
Published 2023
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:In this study, prompt engineering that improves prompts was explored to improve the performance of teaching and learning materials recommendations using generative artificial intelligence such as GPT and Stable Diffusion. Picture materials were used as the types of teaching and learning materials. To explore the impact of the prompt composition, a Zero-Shot prompt, a prompt containing learning target grade information, a prompt containing learning goals, and a prompt containing both learning target grades and learning goals were designed to collect responses. The collected responses were embedded using Sentence Transformers, dimensionalized to t-SNE, and visualized, and then the relationship between prompts and responses was explored. In addition, each response was clustered using the k-means clustering algorithm, then the adjacent value of the widest cluster was selected as a representative value, imaged using Stable Diffusion, and evaluated by 30 elementary school teachers according to the criteria for evaluating teaching and learning materials. Thirty teachers judged that three of the four picture materials recommended were of educational value, and two of them could be used for actual classes. The prompt that recommended the most valuable picture material appeared as a prompt containing both the target grade and the learning goal. 본 연구에서는 GPT, Stable Diffusion과 같은 생성형 인공지능을 이용한 교수·학습 자료 추천 성능 향상을 위해 프롬프트를 개선하는 프롬프트 엔지니어링에 대해 탐색하였다. 분석할 교수·학습 자료의 종류는 그림 자료이다. 프롬프트 구성에 따른 영향을 탐색하기 위해 명령만 담긴 Zero-Shot 프롬프트, 학습 대상 학년 정보가 담긴 프롬프트, 학습 목표가 담긴 프롬프트, 학습 대상 학년과 학습 목표가 모두 담긴 프롬프트를 설계하여 각각을 GPT-3.5모델에 입력하고 응답을 수집하였다. 수집한 응답을 Sentence Transformers로 임베딩 하고 t-SNE를 활용하여 차원 축소하여 시각화 한 다음 프롬프트와 응답 간의 관계를 탐색하였다. 그리고 각 응답을 k-means clustering algorithm을 활용하여 군집화 한 다음 가장 넓은 클러스터의 첫 번째 값을 대표로 선택하여 Stable Diffusion을 이용하여 이미지화 한 다음 교수·학습자료 평가 기준에 따라 초등학교 교사 30명에게 평가 받았다. 초등학교 교사 30인은 추천한 4종의 그림 자료 중 3종은 교육적 가치가 있다고 판단하였으며, 그 중 2종은 실제 수업에 사용할 수 있다고 하였다. 가장 가치 있는 그림 자료를 추천한 프롬프트는 대상 학년과 학습 목표가 모두 담긴 프롬프트로 나타났다.
Bibliography:KISTI1.1003/JNL.JAKO202325051055458
ISSN:1598-849X
2383-9945