Figure 4.
Receiver operating characteristic (ROC) curves for the four classification algorithms that we used in this study. The shape of each curve indicates the quality of the corresponding method. The closer an ROC curve is to the diagonal of the plot, the worse is the corresponding method. A hypothetical perfect method would have an ROC curve that is a constant function with true positive value 1 and false positive value 0. All four methods that we tested performed significantly better than baseline (ROC score of 0.5, corresponding to a random-guess method) (see Table 4). The logistic regression performed best.
