Predictive System: Comparison of Classification Techniques for Effective Prediction of Heart Disease

Abstract

Today’s world is challenging to most of its people with major concerns for keeping up a good health. Among these challenges, one of the most haunting ones is heart disease. Worldwide, the maximum number of deaths is related to heart diseases. Most of the affected people suffering from heart-related diseases are unaware of their health conditions, and cases are reported at a very later stage, which becomes challenging for doctors to advise them proper treatment and medication with lifestyle changes. This research work aims in comparing classification techniques in finding out which is the most efficient one to predict the disease in less time. Mining important factors and analyzing the relativity between them help in predicting if the patient is having heart disease. The classification techniques used are SVM, Decision Tree, Naïve Bayes, KNN, Random Forest, Ensemble Classification (Extra Trees) and Logistic Regression.