Figure 2.

ROC curves of the evaluation measures in the D. melanogaster dataset. ROC curves for the evaluation measures in the D. melanogaster dataset. The dashed lines indicate the true/false positive rates of the average optimal cutoff, i.e., the average cutoff value that maximises the difference between true and false positive rates (TPR = TP/(TP + FN), FPR = FP/(FP + TN)). The negative sets consist of 1000 samples, each of the size of the positive set, drawn from the non-overlapping candidates. The average optimal cutoff value and the average AUC (area under the curve) value are also shown.