Deep Dive into AUC in Machine Learning (Area Under the ROC Curve) (Part 1)

This is the first part of a 4-part Series on using the Area under the ROC Curve (AUC, AUROC, or C-statistic) in Machine Learning and Statistics. In this video we discuss why the AUC is important in Machine Learning. Specifically, it measures the performance of a ranking prediction in binary classification models. This is in contrast to most other classification metrics, which measure the quality of a hard prediction. Other metrics require the choice of a threshold to evaluate the model. This is a disadvantage, since it is not always clear where to choose the threshold, and can be difficult to determine if poor performance is due to the choice of threshold. AUC is "nice" in that it "averages" the performance across all distinct thresholds, in a way that will be made clear in Part 2. Part 1:    • Deep Dive into AUC in Machine Learning (Ar...   Part 2:    • Deep Dive into AUC in Machine Learning (Pa...   Part 3:    • Deep Dive Into AUC (Part 3) (Equivalence t...