Figure 6From: An active learning based classification strategy for the minority class problem: application to histopathology annotationPerformance of the PBT Classifier Trained on Different Target Class Ratios. Performance of the PBT classifier trained using training sets with different percentages of samples for which r ↪ ω1. Shown are the (a) accuracy and (b) AUC values for the trained classifier at each iteration, using p0(r ↪ ω1) = 0.04.Back to article page