Diabet Diagnosis with Support Vector Machines and Multi Layer Perceptron

Abstract

Diabet is one of the metabolic trouble which is generally occurs genetic and environmental components. It happens increasing of blood level. In this study, diabet illness has been diagnosed with its features by classification with support vector machines (SVM) and artificial neural networks (multi layer perceptron). The method used for diagnosis is aritificial neural networks multi layer perceptron. We used SVM-Linear, SVM-Polinomial and SVM-Radial models. Diabet data set which will be used in our experiments obtained from UCI web site and organized. In this study, we compared several algorithms to diagnose illness rates. Diagnose right predictions (accuracy) are %77.08 for multi layer perceptron, %77.47 for support vector machines, %55.33 for polynomial kernel, %65.10 for radial based kernel and sigmoid kernel. Maximum recognition rate is %77.47 for SVM learning method.