Figure 1
From: Prediction of essential proteins based on gene expression programming
ROC curves and AUC values of ten classifiers trained from 10-fold cross-validation. Original data are divided into 10 equal datasets, and nine-folds are used to train the classifier and the remaining one fold is used for testing. The process is repeated ten times to generated ten classifiers, with each of the ten datasets used exactly once as testing data. The figure illustrates the ROC curves and corresponding AUC values of these classifiers.