A Comparative Analysis on a Diabetic Disease Prediction Considering Various Machine Learning Algorithms

Diabetes is a serious health concern almost all over the countries as it impaired many organs in human body and rate of affected person is gradually increasing. Several significant factors have been incorporated for prediction of diabetic diseases through the machine learning attribute. In this study a non-communicable disease, diabetes, is predicted based on several machine learning approaches like Support vector machine (SVM), Decision tree (DT), Logistic regression (LR), K-Nearest Neighbours (KNN), Gaussian Naive Byes(GNB),Random Forest(RF). Data set of Sylhet Diabetic Hospital, Sylhet in Bangladesh containing 520 samples is collected from online source. Comparing the classifiers SVM shows the maximum accuracy 99%. Since a diabetic patient suffers immensely throughout the life, diagnosis along with proper medication is important.