Development of an Explainable Machine Learning Model for Automatic Heart Failure Prediction

This research investigates the application of machine learning techniques to improve heart failure survival rates. By analyzing a publicly available dataset, the study identifies key features that contribute to heart failure prediction. A variety of machine learning classifiers were employed, including XG Boost, K-nearest neighbors, Naive Bayes, Random Forest Classifier, Logistic Regression, and Support Vector Classifier. Among these, XG Boost demonstrated the highest accuracy of 87%. The findings of this research highlight the potential of machine learning in addressing the challenges associated with heart failure prediction and diagnosis. By optimizing hyperparameters and selecting appropriate machine learning algorithms, significant improvements in prediction accuracy can be achieved. This study contributes to the advancement of healthcare by providing valuable insights into heart failure prediction and informing the development of more effective prevention and treatment strategies.