Prediction of Heart Disease Using Multi-Layer Perceptron Neural Network and Support Vector Machine

In recent years, heart disease is one of the major causes of death. So it is necessary to design a system that correctly diagnoses heart disease. In this study, we have proposed two classifiers. One is a Multi-Layer Perceptron neural network (MLP) and another is Support Vector Machine (SVM). Our work is to classify two-class of heart disease and five class of heart disease. Here we have used the Cleveland heart disease online database which consists of 303 instances with 5 classes and 13 attributes. For two-class classification problem, SVM has 92.45% accuracy while the accuracy of MLP is 90.57%. For five-class classification problem, MLP has an accuracy 68.86% while SVM is 59.01%.