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

Figure 5

External validation performance. External classifier performance. The histogram shows the mean testing performance when classifiers are validated in test sets based on similar platforms or different platforms as from which they were developed. Each bar represents the mean balanced accuracy by 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 (VOTE), respectively.