• P.Pavithra
• Dr.P.B.Pankajavalli

In this emerging world, diseases are increased besides treatment for those diseases also improved. At the same time, diagnosis of disease can lead more time consumption for experts because of its wide range of data. The data mining is the effective technology which can handle numerous amounts of data and extract hidden information from it. It is the emerging area in the field of medical diagnosis.Diabetes also referred as diabetes mellitus, defines group of metabolic diseases in which the person experiences high blood sugar level, because shorter insulin production or body cell’s improper response to the insulin. Three major types of diabetes are type 1 diabetes, type 2diabetes and gestational diabetes. Among various data mining approaches, classification algorithms perform more efficient for the diagnosis of disease. In this paper, we have considered 10 classification algorithms like Naïve Bayes, Multilayer Perceptron, Adaboost, Logiboost, Decision Table, OneR, PART, J48, Random Forest, Random Tree, REPT and comparative analysis are performed. Finally, the best algorithm for diabetes diagnosis is keyed out based on its accuracy and error rate.

Diabetes Mellitus, Naïve Bayes, Multilayer Perceptron, Adaboost, Logiboost, Decision Table, OneR, PART, J48, Random Forest, Random Tree, REPT

"A Comparative Analysis and Performance Evaluation of Classification Algorithm to Predict Diabetes Mellitus", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 7, page no.15-21, July-2018, Available :http://www.jetir.org/papers/JETIR1807723.pdf

"A Comparative Analysis and Performance Evaluation of Classification Algorithm to Predict Diabetes Mellitus", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 7, page no. pp15-21, July-2018, Available at : http://www.jetir.org/papers/JETIR1807723.pdf