Klasifikasi Risiko Penyakit Jantung Menggunakan Random Forest: Pengembangan Model Prediktif Melalui Hyperparameter Tuning

The serious global health problem of heart disease places a heavy burden on the world's healthcare systems. Effective prevention and management depend on early identification and accurate risk assessment. In this study, the Random Forest algorithm and Grid Search CV hyperparameter tuning are used to anticipate the risk of developing heart disease. The main data source is the sizable "Heart Disease Dataset" from Kaggle. The goal is to increase the reliability and accuracy of the Random Forest model, enabling healthcare professionals to make defensible decisions based on consistent risk predictions. In order to manage missing values, remove unused columns, and standardize the dataset using Min Max scaling, data preparation is necessary. The initial model accuracy is 98.5%, however following hyperparameter tuning using a 5-fold cross-validated Grid Search CV, this accuracy rises noticeably to an amazing 100%. These findings demonstrate the Random Forest model's improved clinical decision-making capabilities and enhanced prediction efficacy. The study underlines how the Random Forest algorithm, together with hyperparameter tuning, has the potential to revolutionize the healthcare industry and increases our understanding of heart disease risk appraisal. Healthcare professionals may identify high-risk people and put the right preventative measures in place by providing a trustworthy, explicable prediction model. The study's ultimate goal is to lessen the worldwide burden of heart disease and improve the overall health effects for those who are impacted.