Table 3.
Mean and standard deviation (in brackets) of the AUC obtained on the test set by the five classifiers following the fivefold cross-validation, when all the features (ALL) and only the features selected by CFS (SEL) are used
| TZ | PZ | |||||
|---|---|---|---|---|---|---|
| ALL | SELTZ | p value | ALL | SELPZ | p value | |
| LR | 0.65 (0.068) | 0.73 (0.004) | < 0.0001 | 0.80 (0.020) | 0.83 (0.028) | < 0.0001 |
| FFNN | 0.62 (0.084) | 0.61 (0.081) | 0.2713 | 0.77 (0.033) | 0.80 (0.032) | < 0.0001 |
| SVM | 0.43 (0.064) | 0.42 (0.069) | 0.2431 | 0.72 (0.035) | 0.73 (0.028) | 0.0431 |
| NB | 0.73 (0.060) | 0.75 (0.047) | < 0.0001 | 0.78 (0.022) | 0.81 (0.018) | < 0.0001 |
| RF | 0.53 (0.061) | 0.53 (0.071) | 0.32983 | 0.80 (0.023) | 0.80 (0.024) | 0.3272 |
TZ, transition zone; PZ, peripheral zone; ALL, all the features; SELTZ, features selected by CFS for TZ; SELPZ, features selected by CFS for PZ; LR, linear regression; FFNN, feed-forward neural network; SVM, support vector machine; NB, naïve Bayes; RF, random forest