Heart Disease Prediction Using Various Machine Learning Algorithms

Abstract

AbstractCardiovascular disease is a major problem in modern society. The World Health Organization claims that cardiovascular diseases take the most lives when compared to any other disease taking an estimated 17.9 million lives each year. 4 in 5 deaths are due to strokes and heart attacks. Recently, data mining has proved to be an effective technique to analyze large amounts of data and provide valuable insights into it. When combined with machine learning techniques it can help to predict different heart diseases, therefore, helping in easy diagnosis and hence early treatment. In the course of this paper, various state-of-the-art machine learning models were used to achieve the best performance possible for predicting heart disease. The algorithms used are Random Forest classifier, Support Vector Machine classifier, Logistic regression classifier, and K nearest neighbor classifier. The best performance results were achieved by Logistic Regression when we obtained an accuracy of 91.80%. This approach can help to improve heart disease detection.KeywordsHeart disease detectionMachine learningFeature selection