False positive rate

The proportion of actual negative examples for which the model mistakenly predicted the positive class. The following formula calculates the false positive rate:¹

$$\frac{\text{incorrectly classified actual negatives}}{\text{all actual negatives}}$$

which means:

$$\text{FPR} = \frac{FP}{FP + TN}$$

The false positive rate is the x-axis in an ROC curve.

When (not) to use

Use when false positives are more expensive than false negatives.²

In an imbalanced dataset where the number of actual negatives is very, very low, say 1-2 examples in total, FPR is less meaningful and less useful as a metric.²

See also

• Recall (a.k.a. probability of detection, where the FPR is known as probability of false alarm)
• Accuracy
• Precision

Footnotes

1. developers.google.com/machine-learning/glossary#FP_rate ↩

2. ML crash course - Classification ↩ ↩²