Implementasi Random Forest pada Klasifikasi Penyakit Kardiovaskular dengan Hyperparameter Tuning Grid Search

Abstrak

Cardiovascular disease has the potential to cause death if not treated right, because it interferes with the function of the heart. Machine Learning algorithm can be used to do early diagnosis of cardiovascular disease to lower the risk of death. In this study, the classification of cardiovascular disease uses the Random Forest algorithm to determine whether a person has cardiovascular disease or not. Grid Search is also used to do hyperparameter tuning to find the optimal hyperparameter for the Random Forest algorithm. The performance results of the classification model using Random Forest with Grid Search are 73.06% in accuracy, 75.15% in precision, 68.72% in recall, and 71.79% in f1-score.