Akademik Veri Yönetim Sistemi
KARADENIZ 19th INTERNATIONAL CONFERENCE ON APPLIED SCIENCES, Rize, Türkiye, 28 - 30 Kasım 2025, ss.80-87, (Tam Metin Bildiri)
Heart disease is one of the most common causes of death worldwide. Making early diagnosis and risk assessment critically important. Traditional diagnostic methods require invasive tests or are costly. Therefore machine learning based methods that can predict heart disease risk using clinical data are gaining importance. In this study, Support Vector Machines (SVM) optimized with the Artificial Bee Colony (ABC) algorithm were used to predict heart disease risk in individuals. The dataset used in the study contains 11 different clinical characteristics (age, gender, chest pain type, resting blood pressure, cholesterol level, fasting blood sugar, resting ECG results, maximum heart rate, exercise angina, ST depression, and ST segment slope) for 918 individuals. The dataset was divided into three groups: 60% training, 20% validation, and 20% testing. The ABC algorithm was used to optimize the hyperparameters (C and Sigma) of the SVM. The proposed SVM model was trained using a Radial Basis Function (RBF) kernel function and performed binary classification (normal/heart disease risk). The results demonstrate that the SVM model optimized with the ABC algorithm is an effective method for predicting heart disease risk with high accuracy, precision, and recall values.