Diabetes means blood sugar is above desired level on a sustained basis. Diabetes has become a modern day life style disease affecting millions of people around the world. The prime objective of this research work is to provide a better classification of diabetes. There are already several existing method, which have been implemented for the classification of diabetes dataset. In medical sector, the classifications systems have been widely used to exploit the patient’s data and make the predictive models or build set of rules. Data mining is growing in relevance to solving real world problems and hence this can be applied to the diabetes problem as well. The study proposes to use the UCI repository dataset called PIMA Indians Diabetes dataset and decision tree algorithms like C4.5, J48, ID3 and NBs etc. The comparison study includes parameters like sensitivity, accuracy, specificity and features or nodes selected. This hybrid model enables to accurately classify the diabetes dataset and help the people providing treatment as well as those suffering from the disease.