What is an ROC curve?

The ROC curve (Receiver Operating Characteristic) is a graph that shows the relationship between the sensitivity (True Positive Rate) and specificity (1 - False Positive Rate) of a classifier at various decision thresholds. It is used to evaluate the quality of binary classification models. The closer the curve is to the top-left corner, the better the model distinguishes between classes.

Example: if you have a model that predicts the presence of a disease, the ROC curve helps to understand how well the model balances between detecting sick patients (True Positives) and false alarms (False Positives).