Table 5 Performance of supervised learning models for the binary fatigue detection approach.
Rank | Classifier | Accuracy (%) | Precision (%) | Recall (%) | F1-Score (%) |
|---|---|---|---|---|---|
1 | KNN | 84 ± 0.022 | 82 ± 0.022 | 81 ± 0.022 | 81 ± 0.022 |
2 | Random Forest | 81 ± 0.009 | 88 ± 0.009 | 72 ± 0.009 | 75 ± 0.009 |
3 | SVM (linear) | 77 ± 0.024 | 74 ± 0.024 | 72 ± 0.024 | 73 ± 0.024 |
4 | SVM (RBF) | 67 ± 0.004 | 84 ± 0.004 | 52 ± 0.004 | 43 ± 0.004 |
5 | Decision Tree | 63 ± 0.025 | 58 ± 0.025 | 58 ± 0.025 | 58 ± 0.025 |