Skip to main content

Table 3 Averaged AUC values for comparison of different methods

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

Methods      
Datasets Linear kernel Quadratic kernel RBF kernel Hadamard kernel Correlation kernel
GSE1872 0.3788 ± 0.1019 0.3686 ± 0.1136 0.2117 ± 0.0584 1.000 ± 0.000 0.9989 ± 0.0018
GSE32394 0.9456 ± 0.0312 0.5544 ± 0.1248 0.9344 ± 0.0254 0.9589 ± 0.0166 0.9233 ± 0.0294
GSE59246 0.8977 ± 0.0172 0.5386 ± 0.0579 0.8431 ± 0.0379 0.9022 ± 0.0145 0.8562 ± 0.0113
GSE59993 0.8283 ± 0.0226 0.5935 ± 0.0694 0.8347 ± 0.0182 0.8855 ± 0.0088 0.7869 ± 0.0144
GSE25055 0.8575 ± 0.0182 0.4743 ± 0.0393 0.8196 ± 0.0203 0.8653 ± 0.0171 0.7654 ± 0.0152
GSE1379 0.6205 ± 0.0481 0.5237 ± 0.0701 0.6743 ± 0.0427 0.7300 ± 0.0375 0.6419 ± 0.0453
  1. The bold face represents the best performance detected for different compared methods