In recent years, there has been an explosion in the rate of using technology that helps in discovering the diseases. It contributes to treating several complications such as nerve and blood vessel damage, heart problems, and a higher risk of kidney malfunctioning. Data Mining, being the foremost analyzing technique used by researchers that provides effective results in an early diagnosis of diabetes. This research paper focuses on the approaches, namely Decision tree, Naive Bayes and Neural-Fuzzy classifier in predicting disease and their performance is measured using Accuracy evaluation metric. The classification accuracy and response time were compared to the methods using Accuracy and Running Time as Performance criteria. From the study, it is observed that the Decision tree algorithm gives a better accuracy in the overall performance of the classifier among the algorithms under consideration.