@article{ART003338958},
author={Myeong-Sang Kim and Seong-Hyun Park},
title={A Proposal for Problem Comprehension and Learning Assessment through a Hybrid AI-Supported Educational Model},
journal={Journal of The Korea Society of Computer and Information},
issn={1598-849X},
year={2026},
volume={31},
number={5},
pages={317-324}
TY - JOUR
AU - Myeong-Sang Kim
AU - Seong-Hyun Park
TI - A Proposal for Problem Comprehension and Learning Assessment through a Hybrid AI-Supported Educational Model
JO - Journal of The Korea Society of Computer and Information
PY - 2026
VL - 31
IS - 5
PB - The Korean Society Of Computer And Information
SP - 317
EP - 324
SN - 1598-849X
AB - As academic and professional fields become increasingly segmented and specialized, fragmented and individualized problems are on the rise. Consequently, there is a growing recognition within educational curricula of the need to evaluate problem comprehension and the overall problem-solving process.
However, as long as conventional assessment systems remain unchanged, learners are likely to prioritize score improvement through simple answer derivation. Furthermore, recently introduced generative AI models often provide immediate answers, which deprives learners of opportunities for critical thinking and lacks reliability in their educational foundations. This paper proposes an AI model applicable to both self-directed learning and educational assessment. Through fine-tuning, the proposed model establishes a shared academic direction and specialized knowledge as a dedicated educational language model. It also utilizes Retrieval-Augmented Generation (RAG) technology to secure a reliable educational foundation based on the national curriculum and official textbooks. Subsequently, by integrating Dynamic Student State (DSS) analysis with prompting techniques, the model identifies the learner's real-time context and proficiency level. Instead of providing direct answers, it executes a scaffolding strategy that offers step-by-step hints and metacognition-inducing prompts. Ultimately, this approach aims to strengthen learners' self-directed competencies and alleviate the repetitive workload of teachers, thereby fostering an environment where core educational expertise can be fully exercised.
KW - AI;Fine-tuning;RAG;Prompting;Education;Self-Directed
DO -
UR -
ER -
Myeong-Sang Kim and Seong-Hyun Park. (2026). A Proposal for Problem Comprehension and Learning Assessment through a Hybrid AI-Supported Educational Model. Journal of The Korea Society of Computer and Information, 31(5), 317-324.
Myeong-Sang Kim and Seong-Hyun Park. 2026, "A Proposal for Problem Comprehension and Learning Assessment through a Hybrid AI-Supported Educational Model", Journal of The Korea Society of Computer and Information, vol.31, no.5 pp.317-324.
Myeong-Sang Kim, Seong-Hyun Park "A Proposal for Problem Comprehension and Learning Assessment through a Hybrid AI-Supported Educational Model" Journal of The Korea Society of Computer and Information 31.5 pp.317-324 (2026) : 317.
Myeong-Sang Kim, Seong-Hyun Park. A Proposal for Problem Comprehension and Learning Assessment through a Hybrid AI-Supported Educational Model. 2026; 31(5), 317-324.
Myeong-Sang Kim and Seong-Hyun Park. "A Proposal for Problem Comprehension and Learning Assessment through a Hybrid AI-Supported Educational Model" Journal of The Korea Society of Computer and Information 31, no.5 (2026) : 317-324.
Myeong-Sang Kim; Seong-Hyun Park. A Proposal for Problem Comprehension and Learning Assessment through a Hybrid AI-Supported Educational Model. Journal of The Korea Society of Computer and Information, 31(5), 317-324.
Myeong-Sang Kim; Seong-Hyun Park. A Proposal for Problem Comprehension and Learning Assessment through a Hybrid AI-Supported Educational Model. Journal of The Korea Society of Computer and Information. 2026; 31(5) 317-324.
Myeong-Sang Kim, Seong-Hyun Park. A Proposal for Problem Comprehension and Learning Assessment through a Hybrid AI-Supported Educational Model. 2026; 31(5), 317-324.
Myeong-Sang Kim and Seong-Hyun Park. "A Proposal for Problem Comprehension and Learning Assessment through a Hybrid AI-Supported Educational Model" Journal of The Korea Society of Computer and Information 31, no.5 (2026) : 317-324.
