<![CDATA[Cardiovascular disease is one of the most serious health problems and remains the leading cause of death worldwide each year. This condition shows that early diagnosis and classification of cardiovascular disease are very important to increase the chances of recovery and prolong the life expectancy of patients. This study aims to develop a cardiovascular disease classification system using the Support Vector Machine (SVM) method and analyze the effect of Recursive Feature Elimination (RFE) and Chi-Square feature selection on model computational performance. The dataset used is the Heart Disease Dataset from IEEE Dataport, which consists of 1,190 data points. The RFE and Chi-Square methods were used to select the most relevant features, resulting in seven main features used in the classification process with various SVM kernels. The results showed that the application of feature selection significantly improved model stability and accelerated the computational process, especially for kernels with high complexity. The linear kernel provided the highest accuracy of 85.84%, while the seventh-degree polynomial kernel increased from 77.23% to 84.31%, and the RBF kernel increased from 83.66% to 84.86% after feature selection. The combination of the RFE and Chi-Square methods proved to be effective in improving stability and computational efficiency, making it a potential alternative to support the cardiovascular disease classification process]]>
Copyrights © 2026
