In modern day, Diabetes has become one of prominent disease that affecting people all around the world. Over 425 million people worldwide with a majority of them are adults living in low and middle income countries had been affected by diabetes. According to the Diabetes Atlas from International Diabetes Federation in 2017, the number of people affected by diabetes disease are expected to increase to over 629 million by 2045. Along with the increasing of the computer science and information technology in researches, its has continued advent into many field including the medical and healthcare field which the cases of this disease and the symptoms are well recorded and documented. This paper aims to use several data mining techniques such as Naive Bayes, Sequential Minimal Optimization (SMO), RepTree and Simple Logistic Regression for classifying whether positive or negative result of diabetes diagnostic. Then, the results will be measured using confusion matrix in term of accuracy, precision and recall as evaluation metric for measuring the classification performance.
