Table 3.
Performance metrics of machine learning algorithms for Testing data set (n = 20).
| n = 20 | Overall statistics | Statistics by class | Sensitivity | Specificity | Precision | Recall | F1 | Prevalence | Detection rate | Detection prevalence | Balanced accuracy |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Test_CART | Accuracy : 1.000 (0.832, 1.000) | Class: HS | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 | 0.150 | 0.150 | 0.150 | 1.000 |
| Test_CART | P-value [Acc > NIR]: < 0.001 | Class: N | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 | 0.550 | 0.550 | 0.550 | 1.000 |
| Test_CART | Kappa: 1.000 | Class: S | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 | 0.300 | 0.300 | 0.300 | 1.000 |
| Test_CTREE | Accuracy: 1.000 (0.832, 1.000) | Class: HS | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 | 0.150 | 0.150 | 0.150 | 1.000 |
| Test_CTREE | P-value [Acc > NIR]: < 0.001 | Class: N | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 | 0.550 | 0.550 | 0.550 | 1.000 |
| Test_CTREE | Kappa: 1.000 | Class: S | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 | 0.300 | 0.300 | 0.300 | 1.000 |
| Test_KNN | Accuracy: 0.850 (0.621, 0.968) | Class: HS | 0.000 | 1.000 | NA | 0.000 | NA | 0.150 | 0.000 | 0.000 | 0.500 |
| Test_KNN | P-value [Acc > NIR]: 0.005 | Class: N | 1.000 | 0.889 | 0.917 | 1.000 | 0.957 | 0.550 | 0.550 | 0.600 | 0.944 |
| Test_KNN | Kappa: 0.727 | Class: S | 1.000 | 0.857 | 0.750 | 1.000 | 0.857 | 0.300 | 0.300 | 0.400 | 0.929 |
| Test_LDA | Accuracy: 0.900 (0.683, 0.9877) | Class: HS | 0.667 | 1.000 | 1.000 | 0.667 | 0.800 | 0.150 | 0.100 | 0.100 | 0.833 |
| Test_LDA | P-value [Acc > NIR]: < 0.001 | Class: N | 1.000 | 0.889 | 0.917 | 1.000 | 0.957 | 0.550 | 0.550 | 0.600 | 0.944 |
| Test_LDA | Kappa: 0.823 | Class: S | 0.833 | 0.929 | 0.833 | 0.833 | 0.833 | 0.300 | 0.250 | 0.300 | 0.881 |
| Test_RF | Accuracy: 1.000 (0.832, 1.000) | Class: HS | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 | 0.150 | 0.150 | 0.150 | 1.000 |
| Test_RF | P-value [Acc > NIR]: < 0.001 | Class: N | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 | 0.550 | 0.550 | 0.550 | 1.000 |
| Test_RF | Kappa: 1.000 | Class: S | 1.000 | 1.000 | 1.000 | 1.000 | 1.000 | 0.300 | 0.300 | 0.300 | 1.000 |
| Test_SVM | Accuracy: 0.900 (0.683, 0.9877) | Class: HS | 0.667 | 1.000 | 1.000 | 0.667 | 0.800 | 0.150 | 0.100 | 0.100 | 0.833 |
| Test_SVM | P-value [Acc > NIR]: < 0.001 | Class: N | 1.000 | 0.889 | 0.917 | 1.000 | 0.957 | 0.550 | 0.550 | 0.600 | 0.944 |
| Test_SVM | Kappa: 0.823 | Class: S | 0.833 | 0.929 | 0.833 | 0.833 | 0.833 | 0.300 | 0.250 | 0.300 | 0.881 |
No information rate: 0.550.