a
| I | II | III | IV | V |
|---|
TP | 49.4 | 49.4 | 49.0 | 49.4 | 49.4 |
FP | 0.0 | 0.2 | 0.0 | 1.4 | 50.6 |
FN | 0.0 | 0.4 | 0.0 | 0.1 | 0.0 |
TN | 50.5 | 50.4 | 0.6 | 49.2 | 0.0 |
b
| I | II | III | IV | V |
Sn |
100.0
| 99.2 |
100.0
| 99.9 | 100.0* |
Sp | 99.9 | 99.6 |
100.0
| 97.2 | 0.0 |
Acc | 100.0 | 99.4 |
100.0
| 98.5 | 49.4 |
Mcc | 99.9 | 98.8 |
100.0
| 97.1 | – |
Ppv | 99.9 | 99.6 |
100.0
| 97.2 | 49.4 |
Pc | 99.9 | 98.8 |
100.0
| 97.1 | 49.4 |
F1 |
100.0
| 99.4 |
100.0
| 98.5 | 66.1 |
- I: DAX, II: EIIP, III: Complimentary, IV: Enthalpy, V: Galois
- Panel a: the measurement of methods
- TP: True positive
- FP: False positive
- FN: False negative
- TN: True negative
- Panel b: the evaluation of methods
- Sensitivity, Sn=TP/(TP+FN)
- Specificity, Sp=TN/(TN+FP)
- Accuracy, Acc=(TP+TN)/(TP+FP+FN+TN)
- Matthews correlation coefficient, \(Mcc=TP\times TN - FN\times FP \over {\sqrt {(TP + FN) \times (TN + FP) \times (TP + FP) \times (TN + FN)} }\)
- Positive predictive value, Ppv=TP/(TP+FP)
- Performance coefficient, Pc=TP/(TP+FN+FP)
- F1 score, the harmonic mean of precision and sensitivity, F1=2×TP/(2×TP+FP+FN)
- *: Not eligible for comparison due to training failure
- –: Invalid value
