An Efficient Heart Disease Prediction System Using Modified Firefly Algorithm Based Radial Basis Function with Support Vector Machine

Nowadays Heart Disease is one among the main roots of death in and around countries. Accurately predicting the heart disease is diffi-cult for doctors. Thus, it is obligatory to apply computerized technologies to support doctors for diagnose faster with greater accuracy. An existing work introduced a heart disease diagnosis system which is dependent upon Interval Type-2 Fuzzy Logic System (IT2FLS).
However, the training time of IT2FLS as well as genetic hybrid algorithms is quite gentle. And also it does not achieve high recognition accuracy. To solve this problem the proposed system introduced a modified firefly algorithm and Radial Basis Function based Support Vector Machine (MFA and RBF-SVM). An input dataset encompasses 3 kinds of attributes such as Input, Key and Prediction attributes. After the normalization, an attribute reduction and feature extraction are performed by using FA and Principal Component Analysis
(PCA) respectively. Finally RBF-SVM is classified a features as normal or heart diseases.