12,445 papers · continuously updated · last export: 10 Aug 2026livingmeta.ai
← Browse all papers
AI evidence extraction

KTCF: Actionable Recourse in Knowledge Tracing via Counterfactual Explanations for Education

Woojin Kim, Changkwon Lee, Hyeoncheol Kim · Proceedings of the AAAI Conference on Artificial Intelligence · 2026

AI-generated evidence extraction, verified across multiple analytical personas. Not a substitute for the peer-reviewed original.

5/10
Relevance
0/4
Quality (LMQS)
E
Evidence
1
Citations
15.05
FWCI
Top 10%
Impact

This is an AI analysis. Read the peer-reviewed original at the publisher: https://doi.org/10.1609/aaai.v40i45.41216

Methodology & findings

Study design

Empirical experiment on a large-scale educational dataset comparing the proposed KTCF method against existing knowledge tracing methods using performance metrics.

Main result

The study demonstrates that "KTCF method achieves superior and robust performance over existing methods, with improvements ranging from 5.7% to 34% across metrics." Additionally, the research shows that counterfactual explanations can be converted into educational instructions that help reduce study burden for students.

Reports effect sizes.

Research paradigm

positivist/computational

Author conclusions

The authors conclude that "counterfactuals have the potential to advance the responsible and practical use of AI in education" and that "Future works on XAI for KT may benefit from educationally grounded conceptualization and developing stakeholder-centered methods."

Risk of bias

Dataset-specific performance may not generalize across different educational contexts; Potential selection bias in the large-scale educational dataset composition not described; No mention of control for confounding variables in study design; Qualitative evaluation is not described quantitatively, limiting reproducibility

Open questions raised

  • Need for more educationally grounded conceptualization of counterfactual explanations in knowledge tracing
  • Development of stakeholder-centered methods for XAI in education
  • Expansion of practical applications of AI in educational settings
  • The authors identify the need for educationally grounded conceptualization and the development of stakeholder-centered methods in future work on Explainable AI for Knowledge Tracing.
  • Future works on explainable AI for Knowledge Tracing should focus on educationally grounded conceptualization and developing stakeholder-centered methods.
Data: not_statedCode: not_statedExtracted from: pdfAgreement 69%

Explore related topics

AI Literacy TrainingDoctoral TrainingEmpirical Study

Related papers

← Browse all papers