<![CDATA[Cardiovascular disease is one of the leading causes of death worldwide, with a high mortality rate, especially in developing countries like Indonesia. This highlights the importance of developing systems to identify and detect heart disease at an early stage. In this study, the Support Vector Machine (SVM) algorithm was used to classify cardiovascular diseases by utilizing a dataset consisting of 303 patient records obtained from Kaggle. The dataset was divided into 70% for training and 30% for testing. Before optimization using GridSearchCV, the SVM model achieved an accuracy of 79%, precision of 79%, recall of 73%, and F1-score of 76%. However, after adjusting the hyperparameters with GridSearchCV, the model's accuracy slightly decreased to 77%, with precision remaining at 79%, recall dropping to 66%, and F1-score at 72%. Despite this decline in performance after optimization, the results indicate that although SVM has potential for classifying heart disease, its performance is highly influenced by data quality and the selection of appropriate hyperparameters. Even with the performance decrease postoptimization, the model still provides useful predictions, showing consistent results and a proportional class distribution.]]>
Copyrights © 2025
