Penggunaan Decision Tree dalam Penentuan Faktor yang Mempengaruhi Status Gizi Buruk Balita di Kelurahan Tamamaung

Abstract

This study aims to evaluate the use of Decision Tree algorithms in determining the nutritional status of children based on Posyandu activity reports. Malnutrition poses serious risks for developing children, including weakened immune systems, long-term developmental delays, and high mortality rates. By applying the Decision Tree algorithm to classify the nutritional status of toddlers, this research seeks to identify nutritional status, which can then be addressed by health centers (Puskesmas). Using attributes such as weight (W), age (A), and height (H), aligned with child anthropometric indices, the Decision Tree method will be utilized to determine the factors influencing nutritional status in toddlers. The application of this method will facilitate the identification of at-risk toddlers, enabling timely prevention and intervention. Testing through k-fold cross-validation yielded an accuracy of 79.43%, a recall of 53.1%, and a precision of 76.6%. The results indicate that, out of 350 data points, the most significant factor affecting children's nutritional status is weight.