PREDICTION OF DIABETES MELLITUS USING MACHINE LEARNING TECHNIQUES

Diabetes mellitus is a common disease caused by a set of metabolic ailments where the sugar stages over drawn-out period is very high. It touches diverse organs of the human body which therefore harm a huge number of the body's system, in precise the blood strains and nerves. Early prediction in such disease can be exact and save human life. To achieve the goal, this research work mainly discovers numerous factors associated to this disease using machine learning techniques. Machine learning methods provide effectual outcome to extract knowledge by building predicting models from diagnostic medical datasets together from the diabetic patients. Quarrying knowledge from such data can be valuable to predict diabetic patients. In this research, six popular used machine learning techniques, namely Random Forest (RF), Logistic Regression (LR), Naive Bayes (NB), C4.5 Decision Tree (DT), K-Nearest Neighbor (KNN), and Support Vector Machine (SVM) are compared in order to get outstanding machine learning techniques to forecast diabetic mellitus. Our new outcome shows that Support Vector Machine (SVM) achieved higher accuracy compared to other machine learning techniques.