Future Research Directions for Knowledge Tracing¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The survey closes by noting "there are still a substantial number of problems that require urgent resolution." It names six directions: interpretable KT, including educational theories such as the Rasch model or transfer of knowledge; KT with sparse learning interactions via contrastive learning; KT for subjective, open-ended exercises; KT using students' feedback; KT for general user modeling; and KT with large language models.

Design Implications¶

Context¶

Requirements¶

• Interpretable models, because students often express more concern about the why than the what of a learning decision
• Methods for sparse interaction data, open-ended natural-language answers, and student feedback

Constraints¶

• The survey states it is imperative to safeguard the privacy and security of student data when utilizing LLMs
• The authors caution that model complexity may enhance accuracy but compromise applicability, as users will question the reliability of complex models

Target Learners¶

• Researchers developing knowledge tracing for online learning systems

Target Learning Goals¶

• More accurate and interpretable evaluations of students' knowledge states

Related Strategies¶

• Knowledge Tracing Model Taxonomy
• Deep Learning Knowledge Tracing Models

Examples¶

-

Key Sources¶

• Shuanghong Shen, Qi Liu, Zhenya Huang, Yonghe Zheng, Minghao Yin, Minjuan Wang, and Enhong Chen. (2021). A Survey of Knowledge Tracing: Models, Variants, and Applications. https://arxiv.org/abs/2105.15106