Diabetes Mellitus (DM) is a chronic metabolic disease that arises due to abnormally high levels of blood glucose, known as hyperglycemia. Timely diagnosis and treatment are crucial to mitigate life- threatening risks associated with this condition.Fortunately, advances in technology have enabled earlier detection of the disease.In this study, we have developed a decision support system using machine learning algorithms to aid in DM diagnosis. Our approach begins with a very detailed exploratory data analysis, followed by data pre-processing and feature selection techniques. Important features are selected, and the data is trained using four different machine learning algorithms, including Naïve Bayes, Random Forest, Gradient Boosting and Multilayer Perceptron. Performance metrics were used to evaluate the experimental results of the machine learning algorithms.