Performance Analysis of Classification Methods for Cardio Vascular Disease (CVD)

Abstract

Cardio Vascular Diseases (CVD) are a cluster of diseases of blood vessels and heart, which ranges from a small blood clot to severe heart failure. Machine learning classifiers help to forecast the plausibility of patients subjected to Cardio Vascular Disease (CVD) by analyzing various medical parameters such as heart rate, Cholesterol level, HbA1c, weight, ECG results. This paper focuses on the performance of various machine learning classifiers based on accuracy and execution time over a CVD dataset in predicting Cardio Vascular Disease (CVD). RandomForest, J48, Hoeffding tree, Logistic Model Trees (LMT), and RandomTree classifiers were used in the prediction. In the analysis, Hoeffding tree classifier achieved high accuracy of 85.1852% and execution time of 0.17 s in predicting patients subjected to CVD than the other classifiers under analysis.