Table 2 Hyperparameter settings for the baseline models experimental setup.
Method | Parameter configuration |
|---|---|
Linear regression | No regularization |
Ridge regression | L2 regularization coefficient \(\alpha =0.1\) (optimized via cross-validation) |
KNN | Number of neighbors \(k=5\), Distance metric=Euclidean distance |
SVR | Kernel function=RBF, Regularization parameter \(C=1.0\), Kernel coefficient \(\gamma =0.1\) |
RF | Number of trees=100, Maximum depth=10, Feature subset size=\(\sqrt{p}\) (p is number of features) |
GBDT | Learning rate=0.1, Number of trees=200, Maximum depth=3 |
XGBoost | Learning rate=0.05, Number of trees=250, Maximum depth=4, L2 regularization term \(\lambda =1\) |
