📈 ROC Curve Visualization
Understand classifier performance across all thresholds
ROC Curve
0.00
AUC (Area Under Curve)
0%
TPR (True Positive Rate)
0%
FPR (False Positive Rate)
📚 Understanding ROC Curve
- ROC Curve: Plots TPR (sensitivity) vs FPR at all thresholds
- AUC (Area Under Curve): Overall classifier quality. 1.0 = perfect, 0.5 = random
- TPR: Of actual positives, how many detected (Recall)
- FPR: Of actual negatives, how many incorrectly flagged positive
- Threshold: Adjust to balance TPR and FPR based on your needs
- Perfect Classifier: Top-left corner (TPR=1, FPR=0)
- Random Classifier: Diagonal line (AUC=0.5)