The survey states that deep learning knowledge tracing models exhibit superior performance to Bayesian and logistic models but have significant room for improvement in interpretability.¶

Claim · All claims
Evidence · 1 study · review r2 · q2
Currency · newest study 2021 · checked against its sources 2026-10-20 (automated, 1 of 1 entry) · not reviewed by a person · overdue for a recheck since 2024: tool and model accuracy changes quickly

Subclaims¶

q2 i? The review asserts deep learning KT models exhibit superior performance compared to Bayesian and logistic models while leaving significant room for improvement in interpretability and explainability. → Shuanghong Shen 2021

Evidence¶

Shuanghong Shen 2021¶

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 doi:10.48550/arxiv.2105.15106

q2 · i? · review · r2

Narrative review summary of the fundamental models section, with no pooled comparison by the authors: deep learning models are described as "exhibiting superior performance compared to Bayesian mod- els and logistic models" but lacking interpretability, which the survey says has limited their further applicability.

"It is crucial to emphasize that, despite deep learning models exhibiting superior performance compared to Bayesian mod- els and logistic models, there remains significant room for improvement in their interpretability and explainability."

Studies bearing on this claim¶

How accurately does BKT predict a learner's next answers, compared with other knowledge-tracing models, and do its assumptions and fitted parameters hold up? This table gathers the studies behind this claim and the 18 narrower claims that bear on it: 11 distinct studies (7 design, 1 review, 1 causal, 1 associational, 1 theoretical), q1–q3, 1 of 11 with an effect size coded. One row per study, codes as each entry records them; no effect is averaged and no verdict is drawn. Built by scripts/synthesize_claim.py from eval/synthesis/; a row that looks wrong is fixed on the claim it comes from.

Discussion¶

Related Claims¶

• No Single Knowledge Tracing Model Is Always Best
• The survey reports, citing Lu et al., that xAI-based interpretations of deep knowledge tracing were demonstrated to aid in enhancing the trust of students and teachers. — related
• BKT is the most efficient knowledge tracing approach overall in simulated online mastery learning, though DKT is more efficient for AS and M problems — related
• Among deep learning KT models, simpler DKT performs comparably to complex architectures under limited dialogue data — related
• LT-MKT achieves the best performance across all four datasets and all three evaluation metrics against eleven KT baselines — related
• DKT-Sem, a DKT variant using semantic text embeddings, performs better than existing KT methods on tutoring dialogues, with a smaller margin on the larger MathDial dataset. — related
• The proposed Transformer-based framework achieved first place in the EDM Cup 2023 with an AUC of 78.969% on the private evaluation dataset — related
• Symbolic knowledge injection improves temporal reliability of predictions, yielding lower early- and mid-sequence errors and the lowest prediction inconsistency rates across sequence lengths — related
• Existing KT methods fail to beat a majority-class baseline on the small CoMTA dialogue dataset but perform significantly better on the larger MathDial dataset. — related
• On the ASSISTments 2020 dataset, StanBKT inference methods achieve nearly identical predictive performance while differing mainly in fit time — related
• Responsible-DKT outperforms a neural-symbolic baseline without knowledge injection and a fully data-driven PyTorch DKT in predictive accuracy across training ratios and sequence lengths — a narrower finding that bears on this claim
• The survey reports, citing Qiu et al., that BKT consistently overestimates the accuracy of students' answers when a day or more has elapsed since their previous responses. — related