<![CDATA[The heart is a vital organ responsible for pumping blood throughout the body. Hence, impairments can disrupt blood circulation and are the leading causes of mortality worldwide. World Health Organization (WHO) reported that, in 2021, the mortality rate attributed to heart disease reached a significant number. In Indonesia, the prevalence of heart disease attained 1.5%. Consequently, it is essential to prevent and detect heart disease at an early stage utilizing machine learning technologies. This study aims to develop a heart disease classification model using the naïve Bayes and random forest algorithms through the ensemble voting classifier approach. The data were obtained from Kaggle, comprising 1,000 records with 14 variables, including one classification target. Imbalanced data were handled using the synthetic minority oversampling technique (SMOTE), while feature selection was conducted in consultation with cardiologists to ensure clinical relevance. The model was trained using the naïve Bayes algorithm, random forest, and integration of both through the ensemble voting classifier method, in contrast to previous studies that only compared several algorithms to determine the highest accuracy. The test results showed that the model trained with the ensemble voting classifier yielded the best performance, with an accuracy, precision, recall, and F1 score of 98.28%, 98.41%, 98.41%, and 98.41%, respectively. This study demonstrates that the ensemble voting classifier method provides better accuracy than the individual algorithms. This model falls within the excellent classification category and is expected to contribute to the medical field and support the development of decision-support systems for diagnosing heart disease.]]>
Copyrights © 2025
