Table 3.
Results obtained from applying different methods and classifiers.
| Classifier | Method | Accuracy (%) | Sensitivity (%) | Specificity (%) | MCC |
|---|---|---|---|---|---|
| k-NN (k = 1) | LOSO | 53.37 | 49.62 | 57.12 | 0.0007 |
| s-LOO (1–4) | 42.50 | 30.00 | 55.00 | 0.0015 | |
| s-LOO (2–5) | 52.50 | 45.00 | 60.00 | 0.0005 | |
| s-LOO (3–6) | 50.00 | 55.00 | 45.00 | 0.0000 | |
| s-LOO (all) | 55.00 | 55.00 | 55.00 | 0.1000 | |
| MCFS | 67.50 | 75.00 | 60.00 | 0.3549 | |
| A-MCFS | 70.00 | 80.00 | 60.00 | 0.4082 | |
|
| |||||
| k-NN (k = 3) | LOSO | 54.04 | 53.27 | 54.81 | 0.0008 |
| s-LOO (1–4) | 55.00 | 45.00 | 65.00 | 0.1021 | |
| s-LOO (2–5) | 60.00 | 55.00 | 65.00 | 0.2010 | |
| s-LOO (3–6) | 42.50 | 55.00 | 30.00 | 0.0015 | |
| s-LOO (all) | 55.00 | 55.00 | 55.00 | 0.1000 | |
| MCFS | 65.00 | 60.00 | 70.00 | 0.3015 | |
| A-MCFS | 67.50 | 75.00 | 60.00 | 0.3540 | |
|
| |||||
| k-NN (k = 5) | LOSO | 54.42 | 53.65 | 55.19 | 0.0009 |
| s-LOO (1–4) | 55.00 | 45.00 | 65.00 | 0.1201 | |
| s-LOO (2–5) | 57.50 | 65.00 | 50.00 | 0.1517 | |
| s-LOO (3–6) | 50.00 | 70.00 | 30.00 | 0.0000 | |
| s-LOO (all) | 55.00 | 70.00 | 40.00 | 0.1048 | |
| MCFS | 67.5 | 60.00 | 75.00 | 0.3540 | |
| A-MCFS | 72.50 | 70.00 | 75.00 | 0.4506 | |
|
| |||||
| k-NN (k = 7) | LOSO | 53.94 | 54.04 | 53.85 | 0.0008 |
| s-LOO (1–4) | 65.00 | 55.00 | 75.00 | 0.3062 | |
| s-LOO (2–5) | 62.50 | 60.00 | 65.00 | 0.2503 | |
| s-LOO (3–6) | 42.50 | 65.00 | 20.00 | 0.0017 | |
| s-LOO (all) | 57.50 | 65.00 | 50.00 | 0.1517 | |
| MCFS | 62.5 | 65.00 | 60.00 | 0.2503 | |
| A-MCFS | 77.50 | 80.00 | 75.00 | 0.5507 | |
|
| |||||
| SVM (linear kernel) | LOSO | 52.50 | 52.50 | 52.50 | 0.0006 |
| s-LOO (1–4) | 77.50 | 80.00 | 75.00 | 0.5507 | |
| s-LOO (2–5) | 70.00 | 80.00 | 60.00 | 0.4082 | |
| s-LOO (3–6) | 60.00 | 65.00 | 45.00 | 0.2000 | |
| s-LOO (all) | 67.50 | 70.00 | 65.00 | 0.3504 | |
| MCFS | 75.00 | 75.00 | 75.00 | 0.5000 | |
| A-MCFS | 85.00 | 85.00 | 85.00 | 0.6000 | |
|
| |||||
| SVM (RBF kernel) | LOSO | 55.00 | 60.00 | 50.00 | 0.1005 |
| s-LOO (1–4) | 65.00 | 60.00 | 70.00 | 0.3015 | |
| s-LOO (2–5) | 70.00 | 70.00 | 70.00 | 0.4000 | |
| s-LOO (3–6) | 72.50 | 70.00 | 75.00 | 0.4506 | |
| s-LOO (all) | 65.00 | 70.00 | 60.00 | 0.3015 | |
| MCFS | 75.00 | 80.00 | 70.00 | 0.5025 | |
| A-MCFS | 87.50 | 90.00 | 85.00 | 0.7509 | |
|
| |||||
| Naïve Bayes | MCFS | 75.00 | 90.00 | 60.00 | 0.5241 |
| A-MCFS | 80.00 | 80.00 | 80.00 | 0.6000 | |
|
| |||||
| Discriminant analysis | MCFS | 72.50 | 75.00 | 70.00 | 0.4506 |
| A-MCFS | 82.50 | 80.00 | 85.00 | 0.6508 | |
Central tendency metrics used in s-LOO method: 1: mean, 2: median, and 3: trimmed mean (25% removed).
Dispersion metrics used in s-LOO method: 4: standard deviation, 5: mean absolute deviation, and 6: interquartile range.