Skip to main content

Table 2 Averaged AUC values for determining optimal σ in RBF kernel

From: Hadamard Kernel SVM with applications for breast cancer outcome predictions

σ       
Datasets σ=0.01 σ=0.1 σ=1 σ=10 σ=100 σ=1000
GSE1872 0.2379 ± 0.0538 0.2379 ± 0.0538 0.2379 ± 0.0538 0.2379 ± 0.0538 0.2379 ± 0.0538 0.2379 ± 0.0538
GSE32394 0.1811 ± 0.0707 0.1811 ± 0.0707 0.2044 ± 0.0845 0.6767 ± 0.1125 0.9456 ± 0.0133 0.9456 ± 0.0122
GSE59246 0.4408 ± 0.0446 0.4408 ± 0.0446 0.4408 ± 0.0446 0.4408 ± 0.0446 0.8424 ± 0.0379 0.8658 ± 0.0110
GSE59993 0.3542 ± 0.0283 0.3542 ± 0.0283 0.4305 ± 0.0355 0.8392 ± 0.0235 0.6937 ± 0.0340 0.6940 ± 0.0342
GSE25055 0.3651 ± 0.0182 0.3651 ± 0.0182 0.3651 ± 0.0182 0.3651 ± 0.0182 0.8092 ± 0.0156 0.7259 ± 0.0127
GSE1379 0.3952 ± 0.0478 0.3952 ± 0.0478 0.3982 ± 0.0468 0.3970 ± 0.0468 0.6712 ± 0.0294 0.6276 ± 0.0374
  1. The bold face represents best performance detected for different considered σ