Gene Expression Profiles for Predicting Metastasis in Breast Cancer: A Cross-Study Comparison of Classification Methods

Figure 3

Internal validation performance. Shown in blue histograms are the mean 10-times repeated 10-fold cross-validation balanced accuracy performance (bAcc) within the two training datasets: AM and RO. Methods used are random forest (RF), logistic regression (LR), support vector machines with a radial (R-SVM), linear (L-SVM), polynomial (P-SVM), sigmoid kernel (S-SVM), a neural network with a single hidden layer (NNET), or cross-method voting (Voting).