Diabetes Mellitus Prediction Using Ensemble Machine Learning Techniques

Diabetes is a non-communicable disease and currently it is increasing at an alarming rate. It may cause different serious damage in particular; blur vision, myopia, burning extremities, kidney and heart failure. At this moment it is becoming one of the major diseases. Diabetes occurs when the level of sugar crosses a certain level or the human body can not produce sufficient insulin to balance the level. Therefore, diabetes affected patients need to be informed about it so that they can get proper treatments to control diabetes. For this reason, it is important to predict and classify diabetes at an early stage. So, in this analysis, two Machine Learning algorithms have been used to classify diabetes and compared the performances of the algorithms. The collected dataset has 340 instances and each instance has 26 features. In this study, two Ensemble Machine Learning algorithms have been used, namely Bagging and Decorate. Bagging classified the types of diabetes 95.59% accurately, whereas Decorate classified 98.53% accurately.