Bannari Amman Institute of Technology, Erode, India
The information disclosure from clinical datasets is amazingly huge to make a successful clinical conclusion. The objective of the Machine Learning is to gain information from a dataset and adjust it into a fitting structure for additional utilization. Diabetic Mellitus remains as a broadly rising incessant infection, and this is an incredible test around the world. Today, it is basic in different age bunches extent as of youngsters to grown-ups. As the quantity of Diabetic Mellitus persistent have been multiplying each year explicitly in India. In the proposed work, comparative study on various classification algorithm such as Naïve Bayes, Random Forest, Decision tree, K Nearest Neighbor (KNN) , Support Vector Machine(SVM) on dataset to predict whether the given person affected with diabetic or not. In this work, a new ensemble method is identified to provide better accuracy such as 85.44 % compared with existing classification algorithm.