Models can predict best-fit majors from student records, not only pass or fail within a chosen major, enabling cross-program success questions¶

Claim · All claims
Evidence · 1 study · theoretical r3 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly

Subclaims¶

q2 i? Empirically, models already predict best-fit majors from student records, which supports asking where a student is most likely to succeed rather than only whether they will succeed in one program. → Kaushik Dutta 2026

Evidence¶

Kaushik Dutta 2026¶

Kaushik Dutta. (2026). From Precision Medicine to Precision Education: A Vision for AI-Powered Student Digital Twins, Preventive Student Success, and Career-Aligned Academic Pathways. arXiv preprint. https://arxiv.org/abs/2608.06322 doi:10.48550/arxiv.2608.06322

q2 · i? · theoretical · r3

The paper reports, citing Beaulac and Rosenthal and Bettinger and Baker, that random forests predict which major fits a student's profile. This is second-hand evidence attributed by the review; the cited studies' designs are not described here in detail.

"Empirically, models already predict best-fit majors from student records, not only pass or fail within a chosen major (Beaulac and Rosenthal, 2019; Bettinger and Baker, 2014)."

Discussion¶

Related Claims¶

• Baseline LR and XGBoost models achieve comparable predictive performance on OULAD, with XGBoost slightly higher AUC-ROC — related
• Explainable models including logistic regression, kNN and a Bernoulli Mixture Model all reach around 78% accuracy in predicting question-level pass/fail, with well-calibrated probabilities — related