Hybrid genetic algorithms-driven optimization of machine learning models for heart disease prediction

Abstract

Machine learning (ML) models, such as K-Nearest Neighbor (KNN) and Support Vector Machine (SVM), play a vital role in predicting heart disease. However, their performance is often limited by poor hyperparameter selection. This study presents a novel hybrid approach that uses a Genetic Algorithm (GA) to systematically optimize the hyperparameters of KNN and SVM models, leading to improved classification outcomes. The optimization driven by the GA resulted in significant performance improvements, increasing the classification accuracy of KNN to 95.38 % and SVM to 90 %. Furthermore, there were significant improvements in precision, recall, and F-score. Our findings demonstrate that GA-based hyperparameter tuning is an effective strategy for improving the predictive power and clinical relevance of ML models used for heart disease classification.

• •
  GA Driven Optimization: Genetic algorithm was utilized to fine-tune the hyperparameters of K-Nearest Neighbour (KNN) and Support Vector Machine (SVM) classifiers for improved performance.
• •
  Significant Performance Gains: The optimization process aimed to maximize classification across metrics such as accuracy, precision, recall, and F-score.
• •
  Improved Accuracy: The GA-based tuning significantly increased the classification accuracy, improving KNN to 95.38 % and SVM to 90 %.