Prediction of Heart Disease Risk Based on Patient Health History Using the Support Vector Machine (SVM) Algorithm

Septian Simatupang (Politeknik Wilmar Bisnis Indonesia)
Rizki Ramadhansyah (Politeknik Wilmar Bisnis Indonesia)
Rustianna Tumanggor (Universitas Murni Teguh)
Eric Pratama Tan (Politeknik Wilmar Bisnis Indonesia)
Syafrizal Amri Fajar (Politeknik Wilmar Bisnis Indonesia)

Zero : Jurnal Sains, Matematika, dan Terapan
Vol 9, No 2 (2025): Zero: Jurnal Sains Matematika dan Terapan

Heart disease remains the leading cause of death worldwide, with early detection being critical to improving patient outcomes. This study develops a heart disease risk prediction model using the Support Vector Machine (SVM) algorithm. A dataset of 303 patient records with 14 clinical attributes was used, including age, blood pressure, cholesterol, and chest pain type. Data preprocessing, normalization, and feature selection were performed to optimize the model. Evaluation metrics such as accuracy (92%), precision (90%), recall (96%), and F1-score (93%) demonstrated significant improvements over the baseline model. These results highlight the SVM model's effectiveness as a tool for early heart disease detection, offering potential for enhanced predictive healthcare, particularly in Indonesian clinical settings.