Table 1 The six model parameter settings

LR

The glm function was used to train the logistic regression model, and the other parameters were set to the default

SVM

The penalty coefficient of error term was 1, the kernel was linear, the kernel coefficient value was 0.01, the multiclassification decision function was over, the model convergence parameter was 0.001, the maximum number of iterations was 200, and other parameters were set to default values

KNN

K = 5, and all other parameters are set to their default values

RF

Ntree = 500, mtry is the square root of the total number of features, the maximum number of nodes per tree is 20, and the minimum number of samples of leaf nodes is 1

LightGBM

The Metric is "binary_logloss", the number of leaf nodes per tree is 31, the learning rate is 0.1, the number of trees is 100, and the maximum depth of the tree is 30

XGBoost

The evaluation index is logarithmic loss, learning rate is 0.3, maximum depth of the tree is 6, number of trees is 100, subsampling ratio is 1, feature ratio used by each tree is 1, L2 regularization parameter is 1, L1 regularization parameter is 0