Diagnostic Analysis of Diabetes Mellitus Using Machine Learning Approach

Diabetes Mellitus (DM) is caused due to the elevated levels of blood sugar i.e., said be hyperglycemia. The DM a metabolic chronic disease; therefore, early diagnosis and treatment necessary avoid life-threatening risks. According World Health Organization (WHO), diabetes cause high mortality rate with 1.5 million deaths in year. With remarkable improvisations technology, disease can diagnosed earlier. In this paper, we have developed decision-making support machine learning algorithms for diagnosis. Pima Indians dataset was chosen train Machine Learning algorithms. Our approach begins Exploratory data analysis, later sent pre-processing perform feature Selection techniques. important features are selected finally, trained six various (ML) such as Naïve Bayes, KNN, Random Forest, Logistic Regression, Decision Tree, eXtreme gradient boosting. Experimental results ML calculated by performance metrics which that Gradient Boosting has scored highest 88.2% accuracy than other