Building a model for predicting metabolic syndrome using artificial intelligence based on an investigation of whole-genome sequencing

Table 9 Prediction model using feature selecting SNPs

	AUC	Sens	Spec	Prec	F1
Feature selection	randomforest 40 most important SNPs
logistic	0.82	0.634	0.89	0.578	0.605
adaboost	0.81	0.772	0.742	0.415	0.54
Neural network	0.85	0.699	0.834	0.5	0.583

AUC, area under curve; Sens, sensitivity; Spec, specificity; Prec, precision value; F1, F1 score

ISSN: 1479-5876