Applying ML Models for Heart Disease Prediction: A Comparative Analysis

The human body is always susceptible to disease. Heart-related issues are a major concern among them, leading to more annual fatalities than all other diseases combined. It includes various conditions that impact the heart arrhythmias, among others. Alarmingly, heart disease is increasingly prevalent among younger generations, particularly among youth. Research indicates that factors such as chronic stress, smoking, and unhealthy lifestyles contribute to this trend. Given the rising incidence of heart disease, there is a pressing need for advanced algorithms and methodologies for predicting its occurrence. This study aims to determine whether an individual is currently suffering from or is at risk of developing heart disease. Hospitals can leverage automated diagnostic tools that utilize computations for data science and data mining techniques for accurate predictions. This research paper examines the application of supervised ML methods, specifically highlighting the Decision Trees (DT) model, alongside two other models: the Random Forest (RF) model and the LR model. The dataset that was utilized in this investigation came from Kaggle. The Decision Tree and Logistic Regression algorithms demonstrated notable accuracy rates of 88% and 84%, respectively. In contrast, the Random Forest model outperformed the others, achieving an accuracy of 92.74%, almost 93%.