A hybrid prediction model for type 2 diabetes using K-means and decision tree

Type 2 diabetes has a quite high incidence all over the world. For the prevention and treatment of Type 2 diabetes, early detection is demanded. Nowadays, data mining techniques are gaining increasing importance in medical diagnosis field by their classification capability. In this paper, a hybrid prediction model is proposed to help the diagnosis of Type 2 diabetes. In the proposed model, K-means is used for data reduction with J48 decision tree as a classifier for classification. In order to get the experimental result, we used the Pima Indians Diabetes Dataset from UCI Machine Learning Repository. The result shows that the proposed model has reached better accuracy compared to other previous studies that mentioned in the literature. On the basis of the result, it can be proven that the proposed model would be helpful in Type 2 diabetes diagnosis.