Triple voting: hybrid cardiovascular diseases prediction model

Abstract

Currently, cardiovascular diseases are a high-risk cause of death in both developed and developing countries. Thus, heart disease prognosis has received substantial interest in the medical field worldwide. The incidence of heart disorders is escalating at an alarming rate, and it is crucial and worrisome to anticipate their occurrence. Predicting and detecting cardiovascular disease using machine learning and data mining might be clinically useful, but difficult. There are numerous machine learning algorithms accessible, several studies have developed machine learning algorithms for early cardiac disease prediction to help physicians suggest medical treatments. The accuracy of the model will be evaluated to determine whether the performance of the model is accurate or not. Seven machine learning methods are compared in this study, with the data obtained from the UCI Laboratory's cardiovascular patient database. In essence, this research presents a majority voting-based hybrid model which is called triple voting. The hybrid model uses voting of Naïve Bayes, logistic regression (LR) and support vector machines (SVM) experimental outcomes show the proposed triple voting model's accuracy is 89%, which is higher than the individual models and other proposed hybrid models.

Topics & Concepts