Enhancing the Heart Disease Prediction Using Ensemble Voting Classifier

• Published in: 2025 3rd International Conference on Artificial Intelligence and Machine Learning Applications Theme: Healthcare and Internet of Things (AIMLA) pp. 1 - 6
• Main Authors: Sreekanth, G. R., Latha, R.S., Saravanakumar, G., Srikubenthiren, E., Vishal, K., Kavin, B.R.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 29.04.2025
• Subjects: Accuracy; Analytical models; Boosting; Cardiovascular diseases; Classification algorithms; Classification tree analysis; Forestry; Gradient Boosting; Heart; Machine Learning; Measurement; RFE; Vectors; Voting Classifier
• DOI: 10.1109/AIMLA63829.2025.11040390

People who work around this world suffer greatly from a variety of illnesses, and heart disease is particularly prevalent there. Heart disease causes suffering for people of all ages, not just those who are 30 to 60 and older. Nowadays, people under 30 are also suffering because of the unhealthy lifestyle that is prevalent nowadays. This resulted in them dying in their young age. To prevent this, several research projects enhanced the accuracy of heart failure prediction utilizing various machine learning techniques. To enhance prediction performance, this article analyses a variety of ML approaches, such as Random Forest(RF), Logistic Regression(LR), Support vector machine(SVM) and Extra-Boost classifier. Recursive feature elimination (RFE), one of the best feature selection methods, and the gradient boosting algorithm are used to filter the features to obtain the best results from the dataset, which contains 13 attributes. This technique increases the algorithm's performance to an accuracy of 89%. Voting classifier is one of the greatest ensemble techniques for enhancing accuracy further. To achieve accuracy, the Extreme Gradient Boosting classifier(XGB), Random Forest(RF), and Gaussian Naïve Bayes(GNB) are combined using voting classifiers. Finally, the ability to forecast heart disease has increased to an accuracy rate of 98%.