Confidence Bands for ROC Curves: Methods and an Empirical Study

In this paper we study techniques for generating and evaluating confidence bands on ROC curves.  ROC curve evaluation is rapidly becoming a commonly used evaluation metric in machine learning, although evaluating ROC curves has thus far been limited to studying the area under the curve (AUC) or generation of one-dimensional confidence intervals by freezing one variable—the false-positive rate, or threshold on the classification scoring function.  Researchers in the medical field have long been using ROC curves and have many well-studied methods for analyzing such curves, including generating confidence intervals as well as simultaneous confidence bands.  In this paper we introduce these techniques to the machine learning community and show their empirical fitness on the Covertype data set—a standard machine learning benchmark from the UCI repository.  We show how some of these methods work remarkably well, others are too loose, and that existing machine learning methods for generation of 1-dimensional confidence intervals do not translate well to generation of simultaneous bands—their bands are too tight.