Table 12 Parameters used for all methods employed; parameters were selected in a way such that best performance was achieved on the test-set.
From: Outcome prediction based on microarray analysis: a critical perspective on methods
Method Name | Weight Assignment Classifier | Accuracy Measure Classifier | Parameters Used for feature selection | Parameters Used for Classification |
|---|---|---|---|---|
RFE-SVM | SVM | SVM | C = 100 | LK†, C = 100 |
RFE-LNW-GD | LNW-GD | SVM | μ = 10-3, Epochs = 500 | LK†, C = 100 |
RFE-LSSVM | LSSVM | LSSVM | γ = 0.1 | LK, γ = 0.1 |
RFE-RR | RR | SVM | a = 10-1 | LK†, C = 1 |
RFE-FLDA | FLDA | SVM |  | LK†, C = 100 |
RFE-LNW1 | LNW | SVM | μ = 10-2 Epochs* | LK†, C = 100 |
RFE-LNW2 | LNW | SVM | μ = 10-4 Epochs@ | LK†, C = 100 |
RFE-FSVs-7DK | FSV | SVM | 7DK‡, C = 100 | LK†, C = 100 |
Filter | Fisher's Ratio | SVM | - | LK†, C = 100 |