Abstract
Heart disease is a widespread global health concern, leading to significant morbidity and mortality. The timely identification of individuals at high risk of heart disease is crucial for effective prevention and improved health outcomes. Conventional statistical methods have limitations in capturing complex relationships within large datasets of clinical variables. Recent advancements in machine learning offer a promising avenue to improve the accuracy and reliability of heart disease risk prediction. This research leverages machine learning, including logistic regression, decision trees, and Support Vector Machines (SVM), to predict heart disease risk using clinical variables from the Kaggle dataset " Predicting Heart Disease Risk Using Clinical Variables. " The primary objectives are to identify critical predictors of heart disease risk and develop precise predictive models for at-risk patient identification. Such models have the potential to revolutionize heart disease prevention and management strategies.The research paradigm of this study aligns with positivism, emphasizing objective data and empirical evidence. The data collection process is based on purposive sampling, utilizing a publicly available dataset from Kaggle. The analysis involves descriptive statistics , exploratory data analysis, feature selection, and machine learning modeling, including logistic regression, decision trees, and SVM. The results provide insights into the dataset, feature importance, and predictive model performance. This research advances the field of heart disease risk prediction, offering evidence-based insights with implications for health-care professionals and policymakers, ultimately aiming to improve health outcomes for individuals and communities.