Skip to main content

Table 5 Estimation results of supplementing to current screening methods

From: Using machine learning models to improve stroke risk level classification methods of China national stroke screening

Learning methodIncreased number of identited high-risk peopleNumber of misidentited high-risk people
Logistic regression558616,492
Naïve Bayesian493113,977
Bayesian network497113,743
Decision tree(C4.5)564716,097
Neural network560616,208
Random forest548511,722
Bagging with C4.5 decision tree558916,126
Voting538314,536
Boosting with C4.5 decision tree566315,333