Skip to main content

Table 3 Learning algorithms and their default settings.

From: Predictive modeling of structured electronic health records for adverse drug event detection

Classifier Description Notes
DT CART decision tree minimum 1 instance per leaf
SVM Poly Support Vector Machine polynomial kernel of degree 3
SVM RBF Support Vector Machine RBF kernel, gamma = 0.0
LogReg Logistic Regression L2 regularization
kNN k nearest neighbors k = 5
AdaBoost Adaptive boosting Decision trees, 50 base estimators
Bagging Bagging using CART tree 10 base estimators
NB Naïve Bayes  
RF Random forest 500 trees, inspected features = n