Table 2 The Five model parameter settings
From: Establishment of a model for predicting preterm birth based on the machine learning algorithm
Model | Parameter setting |
|---|---|
Logistic Regression Model | Parameters were set to default values |
Decision Tree | The node splitting criterion was Gini diversity index, node partition mode was best, the maximum number of splits was 100, the maximum depth of the tree was unlimited, and other parameters were set to default values |
Naïve Baye | The function was Gaussian radial kernel function (RBF), the smoothing (alpha value) was 1, and other parameters were set to default values |
Support Vector Machine | The penalty coefficient of error term was 1, the kernel was RBF, 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 2000, and other parameters were set to default values |
AdaBoost | The learner type was the decision tree, the maximum number of splits was 20, the number of learners was 30, and the learning rate was 0.1 |