Improved Bio-Inspired Algorithm Design for Prediction of Heart Diseases

Abstract

Optimization methods are used to address dynamic, challenging, and robust problemsto predict heart problems, the vast majority of machine learning algorithms are used. Machine learning for prediction frequently uses classification algorithms as one of its methodologies.Certain categorization techniques forecast accuracy within a reasonable range, while others might not. In this research, Bat and Genetic algorithms are utilized to streamline for various bioinspired algorithms that are used to forecast cardiac disease. Here, we use these bio-inspired algorithms to extract the important elements from the attribute for heart disease. The various classifiers are then updated with these retrieved features. In this study, we look at various bio-inspired algorithm strengthened using classifiers from Random Forest and SVM and compare the results using different metrics, precision, recall, accuracy, etc. Genetic Algorithm is best optimized among these Bio-Inspired Algorithms.