PREDICTION OF DIABETIC DISEASE USING ENSEMBLE CLASSIFIER

In today’s human life, diabetic is the most vulnerable and non-communicable disease creating a great impact in their life. Change in lifestyle and work culture of the people results in millions of diabetic in 21stcentury. Huge amount of data are generated in the modern world, by means of computational analytics on clinical big data. This data are put intocreating a medical intelligence that could be drive the forecasting and prediction. This development in medical intelligence results in great benefit to the people by reducing the hospital re-admission and medical cost, by making this system a patient-centric. Reducing the optimal cost and run time is the result provided by means of improving the health care system by data analytics. In this paper, thediabeticdata aregathered from Kaggle repository. Initially, data has to be pre-processed and randomly divided into training and testing data. Then different ensemble algorithms namely, Bagging, Boosting and Stacking are used for predicting the diabetic disease. Finally ensembleclassifiers results are evaluated by using variousvalidation metrics.