Detection of chronic kidney disease by using ensemble classifiers

Chronic kidney disease is a major health problem that affect the lives of millions of people around the world and causes serious economical, social and medical problems. Chronic kidney disease can be detected with several automatic diagnosis systems. In this study, we apply Adaboost, Bagging and Random Subspaces ensemble learning algorithms for the diagnosis of chronic kidney diseases. Decision tree based classifiers are used in the decision stage. The classification performances are evaluated with kappa and accuracy criteria. Considering the performance analyses of the proposed systems, it is observed that ensemble learning classifiers provide better classification performance than individual classifiers.