AUC-ROC Curves

Definition: AUC-ROC stands for Area Under the Receiver Operating Characteristic Curve. The ROC curve is a graphical plot that illustrates the diagnostic ability of a binary classifier system as its discrimination threshold is varied.

ROC Curve:

• True Positive Rate (TPR): Also known as recall or sensitivity, plotted on the Y-axis.
• False Positive Rate (FPR): Calculated as (FP / (FP + TN)), plotted on the X-axis.

Interpretation: The ROC curve plots TPR against FPR at various threshold settings. The AUC represents the degree or measure of separability. It tells how much the model is capable of distinguishing between classes.

• AUC = 1: Perfect model.
• 0.5 < AUC < 1: Good model.
• AUC = 0.5: No discrimination (random guess).
• AUC < 0.5: Worse than random guess.

Usage: AUC-ROC is particularly useful for comparing the performance of different models. The higher the AUC, the better the model's performance at distinguishing between the positive and negative classes.