The Prediction of Heart Failure based on Four Machine Learning Algorithms

Abstract

The main purpose of this study is to observe the correlations between clinical features (input) with heart disease (output) in the practice of machine learning and modeling. Primarily, this research conducts a data exploration of the attributes and the output to observe the relationships between the attributes and heart disease. The experiment method covers the Decision Tree, K-Nearest Neighbor, Support Vector Machine and Extreme Gradient Boosting algorithms after the data exploration. This research concludes that exercise-induced angina and ST depression are two factors that are highly related to heart disease. Older males should take more care of these features because they are more likely to have heart disease. Moreover, based on the output of four algorithms, the SVM is the best method for predicting the probability of the occurrence of heart disease since SVM output the highest values on the accuracy, recall rate, and f1-score.