Table 2.
Notation | Description | Upper Bound |
---|---|---|
Accuracy | (TP + TN)/(TP + FN + FP + TN) | 1 when FN = 0 and FP = 0 |
Sensitivity (Recall, True positive rate) | TP/(TP + FN) | 1 when FN = 0 |
Specificity (True negative rate) | TN/(FP + TN) | 1 when FP = 0 |
Precision | TP/(TP + FP) | 1 when FP = 0 |
Balanced classification rate | (SN × SP)1/2 | 1 when SN = 1 and SP = 1 |
F1-score | (2 × SN × Precision)/(SN + Precision) | 1 when SN = 1 and Precision = 1 |
TP: true positive; TN: true negative; FP: false positive; FN: false negative; SN: sensitivity; and SP: specificity.