Diabetes mellitus is a chronic disease, lifelong condition that affects the body's ability to use the energy found in food. The level of morbidity and mortality due to diabetes and its potential complications are enormous and pose significant healthcare burdens. It is a complex and time-consuming task in detecting the risk of acquiring diabetes mellitus when a large amount of data is manually processed in the clinical environment. The objective of this paper is to simplify the process of analyzing and detecting the risk of developing diabetes. Patients’ details are gathered and stored in the form of Electronic medical record (EMR). Association rule mining and decision tree induction are applied to the records stored, in order to obtain the set of rules that are to be satisfied. C4.5 or Support vector machine is used to classify the data set accordingly and summarization techniques are used to summarize resultant possibility of acquiring diabetes.