Performance Analysis of Various Data mining Classification Algorithms on Diabetes Heart dataset

Abstract

This Data mining techniques are applied in building software for fast and easy classification models. Early identification has high-risk modules also likely to have a high number of faults. Classification tree models are simple and effective as software quality prediction models while predictions of defects from such models can be used to achieve high software reliability. In this paper, the performance of some data mining classifier algorithms named J48, Random Forest, Random Tree, REP and Naïve Bayesian classifier (NBC) are evaluated based on 10 fold cross validation test. Diabetes is the most rapidly growing chronic disease of our time. People with diabetes are more likely to cause of new blindness, kidney disease, amputation and cardiovascular disease (heart disease and stoke). In this paper we take diabetes and heart datasets relate with their matching fields then apply the classification algorithm in diabetes heart dataset in WEKA (software tool) finding weather people affected by diabetes are getting chance to get heart disease or not, output are evaluated as Tested Negative (No Diabetes), Tested Normal(Not affected), Tested High(affected).