<![CDATA[Heart disease is a serious global health problem, causing 9.4 million deaths each year and is expected to increase to 23.3 million by 2030. The lack of early detection and unhealthy lifestyles are major factors contributing to the rise in cases, especially in developing countries. This study aims to develop an accurate and efficient heart disease risk prediction system as support for early diagnosis. The methods used involve four classification algorithms: Decision Tree, Random Forest, K-Nearest Neighbors (KNN), and Support Vector Machine (SVM), which are evaluated based on accuracy, precision, and F1-score as well as cross-validation. Previous research has shown that KNN is effective in classifying medical data and can detect accurately. In this study, the evaluation results show that Decision Tree and Random Forest have the best accuracy, reaching 99%. Meanwhile, KNN and SVM have accuracies of 84% and 88%. Therefore, the selection of the model must consider the balance between accuracy and generalization ability. It is recommended to use larger and more diverse datasets to improve the reliability of the model in real-world applications, so that early detection systems can help reduce mortality rates due to heart disease.]]>
Copyrights © 2025
