In medical field large amount of data is generated but it is not properly utilized. Many approaches have been taken and examined for effective utilization of data. It uses data analysis tools to find previously unknown interesting pattern from large data sets.

The dataset used is the Pima Indians Diabetes Data Set, which collects the data of patients with and without diabetes.

This paper examines different classification techniques such as Naive Bayesian, Kstar, Random Forest, CART to know which classification technique works better for Pima dataset.