Application of Bagging and Boosting Approaches Using Decision Tree-Based Algorithms in Diabetes Risk Prediction

Abstract

Diabetes is a serious condition that leads to high blood sugar and the prediction of this disease at an early stage is of great importance for reducing the risk of some significant diabetes complications. In this study, bagging and boosting approaches using six different decision tree-based (DTB) classifiers were implemented on experimental data for diabetes prediction. This paper also compares applied individual implementation, bagging, and boosting of DTB classifiers in terms of accuracy rates. The results indicate that the bagging and boosting approaches outperform the individual DTB classifiers, and real Adaptive Boosting (AdaBoost) and bagging using Naive Bayes Tree (NBTree) present the best accuracy score of 98.65%.