Calculate the false negative rate from a confusion matrix and classification results.
The false negative rate (FNR), also called the miss rate, measures how often the model fails to catch a true positive: FNR = FN / (FN + TP) = 1 − sensitivity. It directly quantifies the 'missed detections' side of a classifier's error profile. FNR is especially critical in safety and screening contexts, where a missed positive (e.g. an undetected disease or security threat) can carry a much higher cost than a false alarm.
FNR = FN / (FN + TP)
FNR measures missed true positives (FN / (FN+TP)), the mirror of sensitivity, while FPR measures false alarms among true negatives (FP / (FP+TN)), the mirror of specificity — both quantify different error types on opposite classes.
In domains like disease screening, security threat detection, or safety-critical systems, a missed true positive (high FNR) can have far more severe consequences than a false alarm, so minimizing FNR is often prioritized even at the cost of a higher FPR.
Yes — in classical hypothesis testing terms, FNR corresponds exactly to the Type II error rate (failing to reject a false null hypothesis, or here, missing a true positive).