Table 3 Comparison of h-BeReTa and BeReTa through statistical binary classification tests

True positives (TP)

17

5

False positives (FP)

2

3

False negatives (FN)

10

10

FP/FN

0.2

0.3

Sensitivity or true positive rate (TPR), TP/(TP + FN)

0.63

0.33

Precision or positive predictive value (PPV), TP/(TP + FP)

0.893

0.625

False negative rate (FNR), 1-TPR

0.37

0.67

False discovery rate, 1-PPV

0.105

0.375

F1 score (0 = worst, 1 = best)

0.739

0.435