A random forest-based prediction of cardiovascular diseases

Abstract

With the rapid development of human arithmetic and new algorithms the use of machine learning in the healthcare industry is growing rapidly. A cardiovascular dataset is used to explore the correlation between various data in the dataset and cardiovascular diseases. Five types of machine learning algorithms, namely logistic regression, Adaboost, decision tree classifier, random forest, and neural network, are used to predict cardiovascular diseases. After training, it can efficiently process input samples with high-dimensional features, and the Random Forest model, which integrates multiple trees through the Bagging idea of integrated learning, has the highest score; it evaluates the importance of each feature in the classification problem, and the performance of various algorithms is comprehensively compared using accuracy, precision, recall, and F1-score, which results in the Random Forest as the best model. The use of random forests not only accurately predicts cardiovascular disease, but may also prolong the lives of patients with cardiovascular disease if appropriate methods are used to treat them.