From: AUD-DSS: a decision support system for early detection of patients with alcohol use disorder
Model | Hyper-Parameters |
---|---|
Random forest | Number of trees in the forest = 50, maximum depth of each tree = 20, the minimum number of samples to split each node = 8 |
XGBoost | Learning rate = 0.3, maximum depth of each tree = 6, minimum loss reduction to split each node = 1, regularization term on weights = 20, subsample ratio of columns for each tree = 0.5 |
Decision tree | Maximum depth = 12 |
K-nearest neighbor | Number of k = 7 |
Support vector machine | Kernel = Radius basis function, C = 1, Gamma (γ) = 0.001 |
Logistic regression | Batch size = 100, Debug = True, Standardize attribute = True, Maximum number of iterations to perform = 100, Ridge value in the likelihood = 1.0E-8, conjugate gradient descent = True |