Rule Induction and Prediction of Chronic Kidney Disease Using Boosting Classifiers, Ant-Miner and J48 Decision Tree

Abstract

Chronic Kidney Disease (CKD) is one of the deadliest diseases that slowly damages human kidney. The disease remains undetected in its early stage and the patients can only realize the severity of the disease when it gets advanced. Hence, detecting such disease at earlier stage is a key challenge now. Data mining is a branch of Artificial Intelligence that is widely used to derive interesting patterns from a large volume of medical data. While various data mining techniques used by Experts, boosting and rule extraction techniques have rarely been applied in analyzing Kidney diseases. Boosting is a method of ensemble technique that enhances the prediction power of a data mining model. AdaBoost and LogitBoost are used here for comparing the performance of classification. Ant-Miner is also a data mining algorithm that applies Ant Colony Optimization technique. Ant-Miner along with Decision tree have been used in the paper to derive rules. The aim of this paper is two-fold: analyzing the performance of boosting algorithms for detecting CKD and deriving rules illustrating relationship among the attributes of CKD. The best information retrieved by both classification and rule generation techniques are promising and can be adopted by the Medical Scientists for their research purpose.