Ensemble Classifier for Stroke Prediction with Recurshive Feature Elimination

Abstract

This research proposes an ensemble classifier approach for stroke prediction utilizing Recursive Feature Elimination (RFE). By iteratively selecting and excluding features, RFE enhances the model's predictive capacity while minimizing overfitting. The ensemble classifier, formed by combining diverse base classifiers, capitalizes on their complementary strengths to enhance overall predictive performance. Leveraging a comprehensive dataset, the proposed approach demonstrates superior stroke prediction accuracy compared to individual classifiers, underscoring its potential as an effective tool for early stroke risk assessment.