Area Under the Curve

Area Under the Curve (AUC) is a metric used to evaluate the performance of a predictive model, especially in classification tasks. It measures the total area beneath the model’s ability to distinguish between different classes across all possible threshold levels. A higher AUC value indicates better overall discrimination, meaning the model effectively separates positive cases from negative ones. An AUC of 1.0 signifies perfect distinction, while 0.5 corresponds to random guessing. Essentially, AUC provides a single measure summarizing how well a model can differentiate between outcomes regardless of the specific decision point.