Effective Heart-Disease Prediction by Using Hybrid Machine Learning Technique
Km Shiwangi, Jasminder Kaur Sandhu, Rakesh Sahu
Abstract
With the changing lifestyle habits, a significant proportion of the global population is becoming vulnerable to heart diseases, which have emerged as one of the leading causes of mortality. Healthcare professionals rely heavily on patient data to predict the likelihood of heart disease, as high mortality rates associated with coronary diseases necessitate early intervention. However, analyzing patient data manually is a time-consuming and error-prone process, which necessitates the use of automated prediction systems [1]. This study attempts to determine the best machine learning method for estimating the risk of developing heart disease, which can aid non-specialist doctors or medical technicians in accurately assessing disease risk. The study compares and contrasts six alternative machine learning algorithms: logistic regression, support vector machine, decision tree, k-nearest neighbour, Gaussian naive Bayes, and random forest classifier. Random Forest was found to have the highest prediction accuracy of 95% among all the algorithms evaluated [2].