An offbeat bolstered swarm integrated ensemble learning (BSEL) model for heart disease diagnosis and classification

Abstract

In recent years, cardiovascular disease also known as heart disease has emerged as the top cause of mortality throughout the world. Several different cardiac conditions are included. There are many variables that might increase a person's likelihood of developing heart disease; thus, it is critical to develop reliable, rapid methods of diagnosis and treatment. Therefore, the proposed work aims to develop an effective framework, named as, Bolstered Swarm Integrated Ensemble Learning (BSEL) for heart disease detection. Here, the given medical dataset has been cleaned up, transformed, and normalized using the Linear Interpolation Normalization (LIN) technique. For selecting the best features, the Bolstered-up Beetle Swarm Optimization (BBSO) method is used that removes the irrelevant attributes from the standardized datasets. In addition, the best features are employed in the Weighted Ensemble Classification (WEC) model to establish whether or not a patient has heart disease. The accuracy is improved by using an ensemble classifier with the optimal number of hidden neurons, as determined by the Red Colobuses Monkey Optimization (RCoMO) model. The proposed BSEL prediction model's results are tested and assessed during experiment using a variety of performance measures.