Using all collected physiological features improves exam-grade prediction accuracy to 96% over the 80% achieved by prior EDA-only analyses¶

Claim · All claims
Evidence · 1 study (2 entries) · associational r1 · 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? Prior work using nine engineered EDA-only features reached a maximum accuracy of 80% for the final exam; this study reports obtaining superior results with 96% accuracy by using all useful collected features. → Lala Yamazaki and Ramchandra Rimal 2026 q2 i? The study's stated initiatives include obtaining superior results with 96% accuracy over the prior 80%. → Lala Yamazaki and Ramchandra Rimal 2026 (2)

Evidence¶

Lala Yamazaki and Ramchandra Rimal 2026¶

Lala Yamazaki and Ramchandra Rimal. (2026). Leveraging Physiological Signals to Predict Exam Outcomes with Machine Learning. https://arxiv.org/abs/2606.14960 doi:10.48550/arxiv.2606.14960

q2 · i? · associational · r1

The article describes the benchmark Wearable Exam Stress Dataset analyses, which used only skin electrical conductance from EDA; the review of prior work reports those studies "achieving a maximum accuracy of 80% for the final exam".

"Nine engineered features derived from EDA were then used as input features to predict exam grades, achieving a maximum accuracy of 80% for the final exam [11]."

Lala Yamazaki and Ramchandra Rimal 2026 (2)¶

Lala Yamazaki and Ramchandra Rimal. (2026). Leveraging Physiological Signals to Predict Exam Outcomes with Machine Learning. https://arxiv.org/abs/2606.14960

q2 · i? · associational · r1

The article's introduction lists its research initiatives, including "obtaining superior results with 96% accuracy over 80% previously" by leveraging the full multivariate recordings rather than EDA alone.

"(f) obtaining superior results with 96% accuracy over 80% previously."

Discussion¶

Related Claims¶

• Random forest feature importance identifies EDA, ACC, and TEMP as the top three physiological predictors of exam outcomes — related
• Logistic regression performs poorly (mean accuracy 0.4243) and SVM moderately (0.7374) on physiological exam-outcome prediction — related
• Random forest achieves the highest mean test accuracy (0.9590) of all models compared for predicting exam grade categories from physiological signals — related
• Deep learning models (LSTM, GRU, transformer) achieve high and comparable accuracy (0.90–0.93) on physiological exam-outcome prediction — related