Figure 1 | Scientific Reports

Figure 1

From: Prediction of functional outcomes of schizophrenia with genetic biomarkers using a bagging ensemble machine learning method with feature selection

Figure 1

The schematic illustration of the bagging ensemble machine learning method with feature selection. First, the M5 Prime feature selection algorithm is conducted to find a small subset of biomarkers, which serves as the input to the bagging ensemble machine learning method. The concept of the bagging ensemble machine learning method is to create the multiple versions of a base model by bootstrap reproductions. Then, the ultimate prediction is generated by averaging the predictive performance of the multiple versions. The base model was chosen as linear regression in this study.

Back to article page