The ROC curve is used to evaluate machine learning performance and determine how well a given machine learning classifier is working. The ROC curve is visualized as a graph that shows True Positives vs. False Positives. Provides a visual method for summarizing the performance of binary classifier models across multiple different thresholds by measuring the true positive rate against the false positive rate. If the Area Under the Curve (AUC) is greater than 1, then the model is performing better than chance.
