A Stacked Meta Classifier Approach for Predicting Cardiovascular Diseases

Cardiovascular disease (CVD) is the world's leading cause of mortality and is a condition that affects the heart and blood arteries. The World Health Organization (WHO) reports that more than 32 % of the global population succumbs to heart disease. An innovative machine learning model for forecasting cardiac disease is presented in this study. On datasets from the University of California, Irvine (UCI), the proposed approach is assessed. The aim of this research is to improvise the effectiveness of integrated individual classifiers and to minimize the chances of misclassification. To achieve this, the authors have developed an ensemble approach known as the Stacked Meta Classifier. Feature selection algorithms are utilized to enhance classification accuracy and decrease the execution time of the classification system by selecting the most relevant features. The selected feature set, determined by feature selection algorithms, is used to assess the classifiers' performance. Performance measuring metrics are employed to evaluate the stacking classifier performance. The stacked classifier is a combination of Naive Bayes (NB), Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Random Forest (RF), Linear Discriminant Analysis (LDA) and Decision Tree (DT). The proposed stacked classifier attains an impressive accuracy of 96 percent, surpassing other cutting-edge models by promoting a timely detection of heart disease and can improve survival rates.