Unsupervised time-series clustering identifies low performers more accurately than SVM or decision-tree classifiers trained on two temporal statistics¶

Claim · All claims
Evidence · 1 study · associational r1 · q2
Currency · newest study 2019 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? SVM (RBF kernel) and depth-4 decision-tree classifiers trained on mean binge session length and percentage task completion predicted only 3-4 of 7 low-performance learners, versus 6 of 7 located by the unsupervised clustering. → Peach 2019

Evidence¶

Peach 2019¶

Peach, R. L., Yaliraki, S. N., Lefevre, D. and Barahona, M. (2019). Data-driven unsupervised clustering of online learner behaviour. npj Science of Learning 4:14. https://doi.org/10.1038/s41539-019-0054-0

q2 · i? · associational · r1

Comparison of the unsupervised clustering against an SVM (RBF kernel) and a depth-4 decision tree trained on the two statistical features from Fig. 4. Both classifiers showed poor accuracy: "only 3– 4 out of 7 low-performance learners were accurately predicted".

"For both methods, we find that the accuracy of learner classi fication against performance is low: only 3– 4 out of 7 low-performance learners were accurately predicted."

Discussion¶

Related Claims¶

• Decision trees achieve competitive accuracy with black-box models (LR, RF, SVM), which performed slightly better but lack interpretability — related
• An LSTM-based strategy temporal analytics framework outperforms random forest, SVM, and majority-class baselines for stealth assessment predictive accuracy — related
• All three trained classifiers outperformed the zero-rule baseline (28.4% accuracy) for classifying cognitive engagement in discussion posts — related
• Random forest algorithms yielded the best classification performance in K–8 MMLA studies comparing multiple machine learning models — related
• A random forest classified flagged versus non-flagged word problems with the best accuracy of five tested models (AUC = 0.75) — related