AUC-ROC Interpretation

Hard

machine-learning Stripe PayPal Capital One

Multiple Choice

AUC-ROC is always a reliable metric regardless of class distribution AUC-ROC can be misleadingly optimistic with severe class imbalance; precision-recall AUC is more informative here An AUC of 0.92 means the model correctly classifies 92% of all samples AUC-ROC cannot be computed for imbalanced datasets

Correct!

Incorrect — try again next time!

Upgrade to Pro to unlock solutions and explanations