The false-positive rate (FPR) measures how often a classifier labels actual negative instances as positive. It is the share of false positives among all actual negatives. section 5.2.4
Compute and interpret the metric from the confusion matrix:
- Formula: FPR = FP / (FP + TN), multiplied by 100 for a percentage. section 5.2.4
- Recognize FPR as the inverse of the true negative rate, or specificity. section 5.2.4
- Compare FPR with the false-negative rate because the relative cost of false alarms and missed positives depends on the application. section 5.2.4
Raisin classification
A logistic regression produces 53 false positives and 397 true negatives for the negative class. FPR = 53 / (53 + 397) × 100 = 11.78%, meaning the model incorrectly flags fewer than 12% of actual negatives as positive. section 5.2.4
Table 5.1 Confusion matrix for raisin classification. A confusion matrix stores classification metrics from a regression with actual and predicted binary outcomes.
|
Predicted
|
|||
|---|---|---|---|
|
Kecimen (0)
|
Besni (1)
|
||
|
Actual
|
Kecimen (0)
|
tn
|
fp
|
|
Besni (1)
|
fn
|
tp
|
|
Use the confusion matrix to locate FP and TN before calculating the denominator of the false-positive rate. section 5.2.4
