<![CDATA[Heart disease in Indonesia, especially in the productive age, there is always an increase in the number of cases. The main cause of the increase in the number of heart patients is an unhealthy lifestyle and diet. The increase in patients with heart disease also has an impact on decreasing the standard of living. With this in mind, there is a need for research related to comparing classification methods on heart disease datasets. The dataset obtained is not balanced so that an oversampling technique is needed. The oversampling technique used is SMOTE. This research method uses Support Vector Machine (SVM) and Logistic Regression (LR). In order for this research method to be applied successfully, the data acquisition, data pre-processing and data transformation techniques are used to ensure accurate results. The model evaluation technique used is K-Fold Cross Validation. Based on the results of the analysis, it showed that the data partition using k-fold cross validation without oversampling gets the same accuracy value but the precision value is quite low. Conversely, if using the SMOTE technique, the accuracy value is as good as the precision value. The results of the SVM accuracy value get a value of 91.69%. LR is 91.76%. While the results of the SVM precision value of 57.81% and LR 54.82%. If using the SVM oversampling technique, the score is 75.79% and the LR is 75.84%. Meanwhile, the precision value obtained in SVM is 75.74%. At LR by 74.77%.]]>
