Performance Analysis of Heart Disease Classification for Computer Diagnosis System

Wth the widespread increment in the heart stroke rates at adolescent ages, we have to set up a framework to have the option to distinguish the beginning side effects of a coronary illness to anticipate it. It is impractical for a economic people in developing countries to frequently undergo expensive tests like the ECG and thus there needs to be a crepitation machine learning algorithm in place which is convenient and simultaneously solid, in anticipating the odds of a coronary illness. This can be used to predict the vulnerability of a heart disease for the given symptoms like age, sex, pulse rate etc. Classification is one of the most commonly used tool analyze and classify the data. This paper analyzes the different classifier algorithms such as SVM, KNN and MLP for seer heart disease dataset using WEKA 3.8 software. The performance of the classifiers is evaluated against the parameters like classification accuracy, MCC, Precision, Recall, F-Measure, ROC and so on. SVM Classifier is superior to KNN and MLP and it achieves the highest classification accuracy of 85.9% with minimum false positive of 15.3%.