A Random Forest Classifier embedded in the MLADs achieved an F1-score of 0.82 and AUC of 0.89 for course-completion-risk alerts under five-fold cross-validation¶

Claim · All claims
Evidence · 1 study · design r3 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? The Random Forest Classifier generating early alerts about course completion risk yielded F1 = 0.82 and AUC = 0.89, judged satisfactory for formative use. → Zeynab (Artemis) Mohseni 2026

Evidence¶

Zeynab (Artemis) Mohseni 2026¶

Zeynab (Artemis) Mohseni, Italo Masiello. (2026). Co-designing, developing, and implementing multiple learning analytics dashboards for data-driven decision-making in education: a design-based research approach. Education Tech Research Dev. https://doi.org/10.1007/s11423-025-10577-9

q2 · i? · design · r3

The MLADs' predictive component, a Random Forest Classifier analyzing aggregated attendance, DLM activity and assessment patterns, was trained on 70% of historical data and validated on 30% with five-fold cross-validation, yielding "an F1-score of 0.82" and AUC of 0.89.

"The model was trained on 70% of the available historical data and validated on the remaining 30% using five-fold cross-validation. The evaluation yielded an F1-score of 0.82 and an area under the ROC curve (AUC) of 0.89, which were considered satisfactory for the intended formative use"

Discussion¶

Related Claims¶

• CatBoost achieved the best classification performance among tested algorithms for predicting course completion risk (F-measure .77, accuracy 78%, AUC .87) — reports the opposite
• CatBoost is the best-performing algorithm for predicting programme non-completion, with F1-measure of 94.5±0.6, slightly better than Random Forest — related