Research on Predictive Models for Heart Disease Based on Machine Learning

The heart is the core of the human circulatory system, providing a continuous supply of oxygenated blood to different organs in the body. However, when the heart can no longer effectively pump blood to the other organs, heart disease may develop. Therefore, the effective treatment and early diagnosis of heart disease have become the most significant issue in the medical field. Fortunately, the development of artificial intelligence and big data give birth to machine learning, and some statistical models in machine learning are applied to disease treatment and research. By applying machine learning to heart disease prediction, it can get a more accurate prediction with less time, thereby enhancing the reliability of predictive results. This thesis focuses on comparing diagnostic classifiers using five models in machine learning: logistic regression, k-nearest neighbor classification (KNN), decision tree, support vector machine (SVM), and random forest. The performance of the five models is then analyzed using various indicators, including accuracy, precision, recall, and F1 score. By applying different algorithms in machine learning to heart disease prediction, this research aims to achieve a more precise prediction and timely intervention before the outburst of heart disease, helping to reduce the incidence of heart disease.