Research Article
Gene-Based Multiclass Cancer Diagnosis with Class-Selective Rejections
Algorithm 1
Multiclass SVM minimizing an asymmetric loss function.
1 | 2 | 3 for to do | 4 Using the Training Set / | 5 for to do | 6 Train --SVM on , namely solving the QP (16) | 7 Derive the regularization path for , namely compute the | 8 end | 9 Using the Validation Set / | 10 | 11 | 12 repeat | 13 | 14 = cardinality / | 15 Assign to a decision according to (19) | 16 | 17 | 18 construct the new vector according to the | direction of greatest decrease / | 19 | 20 until is minimum | 21 | 22 | 23 end | 24 index := min | 25 |
|