Table 2 The tuning parameter values of machine learning algorithms
Methods | Hyperparameters | Value for death outcome | Value for relapse outcome |
|---|---|---|---|
ANN | size | 5 | 5 |
weight decay | 0.1 | 0.1 | |
SVM | gamma | 0.061 | 0.074 |
cost | 1 | 1 | |
kernel | radial | radial | |
LS-SVM | tau | 0.0625 | 0.0625 |
sigma | 0.099 | 0.131 | |
kernel | rbfdot | rbfdot | |
NB | laplace | 0 | 0 |
use kernel | TRUE | TRUE | |
adjust | 1 | 1 | |
RF | mtry | 7 | 6 |
ntree | 1000 | 500 | |
DT | minsplit | 6 | 6 |
minbucket | 2 | 2 | |
cp. | 0.02 | 0.01 | |
maxdepth | 4 | 8 | |
Bagging | mfinal | 100 | 50 |
maxdepth | 3 | 3 | |
Boosting | mfinal | 100 | 100 |
coeflearn | Breiman | Freund | |
maxdepth | 3 | 3 |