Improving Heart Disease Prediction through Feature Selection for Multi-Label Classification

Abstract

Heart disease is a leading cause of death worldwide, and early prediction and diagnosis are crucial for effective treatment. In this study, we propose a novel approach for heart disease prediction by using feature selection, XGBoost, Ensemble-Feature-Optimization, and a multi-label classification method. The proposed model aims to improve the accuracy of heart disease prediction by selecting relevant features from the dataset, optimizing the feature ensemble, and applying a multi-label classification method to handle the multiple diseases associated with heart disease. To evaluate the performance of the proposed model, we used two datasets, the Cleveland and Statlog heart disease datasets, which consist of patients with various attributes, such as age, sex, blood pressure, and cholesterol levels. We compared the performance of our proposed model with other machine learning state-of-the-art approaches, using various performance metrics, including accuracy, precision, recall, specificity, and F1-score.The experimental results show that our proposed model outperforms other state-of-the-art approaches in terms of prediction accuracy and other performance metrics. The proposed model achieves an accuracy of 99.5% on the Cleveland dataset and 99.95% on the Statlog dataset. Our approach offers a promising method for heart disease prediction and diagnosis, and the results demonstrate the potential of this approach for improving heart disease treatment and management.