Enhancing Predictive Accuracy: Impact of Feature Selection on Heart Disease Prediction With Machine Learning Models

Abstract This research paper builds upon previous work titled "Prediction of Cardiovascular Diseases using Machine Learning Algorithms" by the same authors. The paper investigates the impact of applying feature selection methods (filter and wrapper) and cross-validation on the efficiency of four models: SVM, Decision Tree, Random Forest, and Neural Network. The findings demonstrate that implementing these techniques improved the efficiency of the models compared to the bare models. This study contributes to the field of cardiovascular disease prediction and provides insights into the importance of feature selection and cross-validation in machine learning modeling.