Skip to main content
Fig. 1 | BMC Medical Informatics and Decision Making

Fig. 1

From: Ensemble machine learning approach for screening of coronary heart disease based on echocardiography and risk factors

Fig. 1

Flowchart of the two-step stacking method. The testing set of size 64, named “Testing”, is used to evaluate the proposed method. The validation set of size 72, named “Validation 0”, is used to train the second-step stacking weights \(w^*_k, k = 1, ...., 10\) in Eq. (4). The rest set of size 288 is randomly divided into a first-step training set (named “Training 1” through “Training 10”) of size 230 and a first-step validation set (named “Validation 1” through “Validation 10”) of size 58 to train the 14 individual classifiers \(c^{(k)}_l, l = 1, ..., 14\) and first-step stacking weights \(w^{(k)}_l, l = 1, ..., 14\) in Eq. (3) for 10 times

Back to article page