Why use PR-AUC for rare positives?

Question: Why is PR-AUC often preferred over ROC-AUC for rare positive-class problems? A) It focuses directly on precision and recall for the positive class B) It ignores the positive class entirely C) It is only used for unsupervised learning D) It always gives a higher score than ROC-AUC Correct: A Explanation: PR-AUC is useful when the positive class is rare because it shows the trade-off between finding positives and avoiding false alarms. It is more sensitive to performance on the minority class than ROC-AUC. Topic: advanced ML / precision-recall / imbalanced data