<![CDATA[Heart disease is a non-communicable disease with a high mortality rate both globally and in Indonesia. According to WHO, around 17.9 million deaths occur each year due to cardiovascular diseases. Early prediction is crucial to reducing mortality and improving life expectancy. This study compares the performance of machine learning algorithms Random Forest Classifier and Support Vector Machine in predicting heart disease. The dataset consists of 5432 medical records from cardiac outpatients at RSUD Kabupaten Bekasi in 2024, with two classes (labeled 1 (heart disease) = 3068 and labeled 0 (non-heart disease) = 2364). Models were developed using the Knowledge Discovery in Databases (KDD) approach. Evaluation results show that the Support Vector Machine model achieved the best performance compared to Random Forest Classifier with 65% accuracy, 70% precision, 68% recall, and 64% f-measure. Cross-validation and ROC analysis also indicated that Support Vector Machine obtained the highest AUC score, ranging from 0.67 to 0.68, which is categorized as poor classification.]]>
Copyrights © 2025
