2022 Fiscal Year Final Research Report
Research on Learner Knowledge Models for Supporting Personalized Learning
Project/Area Number |
18K11597
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Review Section |
Basic Section 62030:Learning support system-related
|
Research Institution | National Institute of Informatics |
Principal Investigator |
Sun Yuan 国立情報学研究所, 情報社会相関研究系, 准教授 (00249939)
|
Project Period (FY) |
2018-04-01 – 2023-03-31
|
Keywords | 認知診断 / Q-matrix / 項目反応データ / 知識追跡モデル / パーソナライズド学習 |
Outline of Final Research Achievements |
In response to growing online learning and personalized learning needs, we aimed to dynamically diagnose learner knowledge state, employing Intelligent Tutoring Systems learning logs. We proposed a comprehensive framework merging learner and domain modeling, inferring domain knowledge structure directly from learner data. Our method enhanced the crucial Q-matrix using a data-driven approach, simultaneously gauging learner knowledge states. Blending cognitive diagnostic and knowledge tracing models, we captured learners' knowledge state and learning process concurrently. Harnessing deep learning, we improved knowledge state representation during performance modeling while maintaining interpretability. The enhanced evaluation accuracy and scalability make our approach adaptable to large-scale online environments, promising broader educational applications than traditional methods, and fostering personalized learning.
|
Free Research Field |
教育心理学
|
Academic Significance and Societal Importance of the Research Achievements |
我々の研究は、認知診断モデルの開発に行列分解法や深層学習を活用するというこれまでにない枠組みを提示した。これにより、教育心理学や心理測定学の領域に新たな知識の構築と進歩を促し、学習者の知識習得状況と学習過程への理解を進展させることが期待できる。また、教育現場において、開発したモデルと手法を活用することで学習者一人ひとりの理解度や学習過程を詳細に把握できれば、個別化された学習支援、教育の個別化を実現できる。特に、オンライン学習環境の普及に伴い大量の学習データが取得されるようになれば、それを有効利用し、教室単位での活用にとどまらず、教育格差の是正や生涯学習の推進などにも繋げうる可能性を有している。
|