2026 Volume 34 Pages 355-367
This research proposes a framework to address Japan's static assessment challenges in tracking student growth. By aligning the Japanese Course of Study with the CASE standard, we found that while Large Language Models (LLMs) can aid in defining pedagogical relationships, a Human-in-the-Loop approach is essential to ensure accuracy. Our prototype system, which integrates a learning map and micro-credentials, effectively visualized students' learning progressions. This work establishes a technical foundation for integrating instruction and evaluation, enabling a more dynamic understanding of student achievement. Furthermore, it offers a replicable model for other educational domains and provides practical insights for developers and educators aiming to implement data-driven instructional support.