Detection of Heart Failure by Using Machine Learning

Abstract

One of the problematic ailments seems to be a heart disease, which impacted a large number of individuals worldwide. Inside the world, one of the hardest difficult tasks is to look after people’s health. Heart illness must be promptly and accurately diagnosed to be treated, particularly in the discipline of cardiovascular. A major cause of a rise in mortality is cardiovascular disease. To address cardiac disorders, healthcare facilities and other institutions provide pricey treatments as well as procedures. Therefore, being capable of identifying cardiac disease during its earliest phases will assist individuals all over the country, especially, with the required precautions before it becomes severe. This information has 11 important features that are needed to conduct the study. Support Vector Machines, Decision Trees, Random Forests, Gradient Boost (GB), Logistic Regression, AdaBoost, XGBoost, and the K-Nearest neighbor approach are some of the supervised methods of machine learning employed in this cardiovascular disease prediction. Additionally, a summary of such algorithms’ performances is provided.