A data science team evaluates two models for fraud detection. Model A has precision=0.92 and recall=0.60. Model B has precision=0.65 and recall=0.95. The business requirement is that missing actual fraud is far more costly than a false alarm. Which model should be selected and why?

Question

Accepted Answer

B. Model B, because it has higher recall and catches more fraud cases. In fraud detection, missing actual fraud (false negative) is more costly than incorrectly flagging a legitimate transaction (false positive). Recall measures the proportion of actual positives correctly identified. Model B's recall of 0.95 means it catches 95% of frauds.

A data science team evaluates two models for fraud detection. Model A has precision=0.92 and recall=0.60. Model B has precision=0.65 and recall=0.95. The business requirement is that missing actual fraud is far more costly than a false alarm. Which model should be selected and why?

Related Questions

Discussion