Akademik Veri Yönetim Sistemi
KARADENIZ CONFERENCE BOOK 19th INTERNATIONAL CONFERENCE ON APPLIED SCIENCES, Rize, Türkiye, 28 - 30 Kasım 2025, ss.73-79, (Tam Metin Bildiri)
Body fat percentage is an important indicator for determining individuals' health status and physical fitness levels. Since traditional body fat measurement methods require expensive equipment or are cumbersome, machine learning based methods that can estimate body fat percentage using anthropometric measurements are gaining importance. In this study, Artificial Neural Networks (ANN) optimized with the Artificial Bee Colony (ABC) algorithm were used to estimate body fat percentage in men. The dataset used in the study contains 14 different anthropometric measurements (age, weight, height, neck circumference, chest circumference, waist circumference, hip circumference, thigh circumference, knee circumference, ankle circumference, biceps circumference, forearm circumference, and wrist circumference) from 252 male individuals. The dataset was divided into three groups: 70% training, 15% validation, and 15% testing. The ABC algorithm was used to optimize the hyperparameters of the ANN. The proposed ANN model has a feedforward network architecture consisting of two hidden layers and was trained using the Levenberg-Marquardt training algorithm. The results show that the ANN model optimized with the ABC algorithm is a reasonable and effective method for predicting body fat percentage.