Table 4 Metrics of test datasets of classifiers.
From: Using machine learning to predict student outcomes for early intervention and formative assessment
Assessment criteria | Random forest | SVM | C5.0 | CART |
|---|---|---|---|---|
Accuracy (%) | 73.7% | 68.8% | 75.4% | 65.5% |
Number of correctly classified samples | 91 | 84 | 92 | 80 |
Number of misclassified samples | 32 | 38 | 30 | 42 |
TP rate | 0.737 | 0.688 | 0.754 | 0.655 |
FP rate | 0.263 | 0.312 | 0.246 | 0.345 |
Precision | 0.727 | 0,701 | 0,794 | 0.761 |
Recall | 0.701 | 0.655 | 0.771 | 0.640 |
F-Score | 0.714 | 0.678 | 0.782 | 0.695 |
Kappa value | 0.472 | 0.377 | 0.50 | 0.312 |
