A false positive occurs when a negative observation is classified as positive. In a confusion matrix, it is the count where the actual label is 0 and the predicted label is 1. section 5.2.4 section 6.3.4
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
|
|
Locate false positives at the intersection of the actual negative row and predicted positive column. section 5.2.4
Use false positives to evaluate classification behavior:
- Count them directly by filtering actual labels equal to 0 and predictions equal to 1. section 5.2.4
- Relate the count to specificity and the false positive rate: FPR = fp / (fp + tn). section 5.2.4
- Adjust the probability cutoff when false positives are more costly than false negatives; a higher cutoff generally changes which cases become positive predictions. section 5.2.3
- Inspect false positives alongside true positives, true negatives, and false negatives rather than relying only on overall accuracy. section 6.3.4
Raisin classification
A logistic regression produced 53 false positives: Kecimen raisins, coded 0, were predicted as Besni, coded 1. With 397 true negatives, the false positive rate was 11.78%. section 5.2.4

























