Fig. 2From: Iterative ensemble feature selection for multiclass classification of imbalanced microarray dataComparison of AUC using KNN and SVM. The y-axis indicates AUC (in percentage, %). The x-axis indicates the number of selected gene signatures. In the legend, “Undersampling” is abbreviated to “US” and “Oversampling” is abbreviated to “OS”. The first and the third column are the experiments using KNN. The second and the fourth column are the experiments using SVMBack to article page