Search

Diabetes mellitus is a metabolic disease that causes high blood sugar. The risk factor of diabetes can be reduced significantly by early precise prediction. Lots of literatures published for diabetes prediction, but nearly all of them proposed frequentist Machine learning algorithm to classify and build models. Besides, their data pre-processing methods are not professional. In this literature, we are proposing MCMC filling missing values and Bayesian variable selection for logistic regression model to classify diabetes patients. It shows great performance (AUC = 0.884, sensitivity = 0.805, specificity = 0.875).