Feature Extraction and Machine Learning Models for Heart Disease Prediction: An Exploratory Analysis

Heart disease is a significant issue in today’s society, responsible for more deaths per year than any other disease. It includes a variety of cardiovascular-related issues, including arrhythmias, embryonic heart abnormalities, and coronary artery disease (CAD). This disease is increasingly common among young people, with studies attributing it to chronic stress, smoking, and poor lifestyle choices. As the number of cases continues to rise, there is a need for advanced algorithms and predictive tools for heart disease. The aim of this study is to assess whether a patient is currently experiencing heart disease or is at risk of developing it.. Hospitals can use automated diagnoses based on data science algorithms and data mining approaches to predict heart disease. This study focuses on implementing supervised ML algorithms such as Decision Trees, Random Forest, KNN, Gaussian Naïve Bayes, Logistic Regression, and SVC. The DT and RF algorithms were shown to have $97 \%$ and $93 \%$ accuracy on the dataset utilized in this analysis, which was gathered via Kaggle.