Table 3 Averaged AUC values for determining the optimal β of Hadamard SVM
From: Heterogeneous multiple kernel learning for breast cancer outcome evaluation
Datasets | β = −1 | β = −0.1 | β = − 0.01 | β = 0.01 | β = 0.1 | β = 1 |
|---|---|---|---|---|---|---|
GSE32394 | 0.9778 ± 0.0222 | 0.9500 ± 0.0278 | 0.9544 ± 0.0433 | 0.9444 ± 0.0444 | 0.9356 ± 0.0356 | 0.9611 ± 0.0389 |
GSE59993 | 0.8063 ± 0.0467 | 0.6904 ± 0.0809 | 0.7055 ± 0.0555 | 0.7137 ± 0.0510 | 0.7113 ± 0.0294 | 0.8661 ± 0.0510 |
GSE1872 | 1.0000 ± 0.0000 | 1.0000 ± 0.0000 | 1.0000 ± 0.0000 | 1.0000 ± 0.0000 | 1.0000 ± 0.0000 | 1.0000 ± 0.0000 |
GSE76260 | 0.8550 ± 0.0220 | 0.8346 ± 0.0533 | 0.8595 ± 0.0126 | 0.8313 ± 0.0389 | 0.8226 ± 0.0706 | 0.7673 ± 0.0310 |
GSE59246 | 0.8996 ± 0.0250 | 0.8994 ± 0.0143 | 0.8666 ± 0.0168 | 0.8564 ± 0.0179 | 0.8888 ± 0.0227 | 0.8969 ± 0.0250 |
BRCA1 | 0.9726 ± 0.0134 | 0.9758 ± 0.0089 | 0.9953 ± 0.0047 | 0.9949 ± 0.0051 | 0.9750 ± 0.0174 | 0.9782 ± 0.0161 |
BRCA2 | 1.0000 ± 0.0000 | 1.0000 ± 0.0000 | 1.0000 ± 0.0000 | 1.0000 ± 0.0000 | 1.0000 ± 0.0000 | 1.0000 ± 0.0000 |
BRCA3 | 1.0000 ± 0.0000 | 1.0000 ± 0.0000 | 1.0000 ± 0.0000 | 1.0000 ± 0.0000 | 1.0000 ± 0.0000 | 1.0000 ± 0.0000 |