<![CDATA[Nutritional status in children under five years of age serves as a key indicator in assessing the overall health, growth, and development of children. Conventionally, nutritional status is determined through manual measurements and interpretation of anthropometric tables, which is time-consuming and prone to human error. With advances in technology, machine learning-based approaches can be used to help classify nutritional status more quickly, objectively, and accurately, thereby supporting decision-making in public health. This study focuses on analyzing and comparing the performance of three machine learning algorithms, namely Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Decision Tree (DT) in classifying the nutritional status of toddlers using anthropometric data that includes variables such as age, gender, weight, and height. In this study, the nutritional status categories classified for the toddler weight dataset include: Severely Underweight, Underweight, Normal, and Overweight. The categories for the height dataset include Severely Stunted, Stunted, Normal, and Tall. The research stages included data preprocessing, data splitting into training and testing, and model performance assessment through accuracy, precision, recall, and F1-score matrices. Based on the evaluation results of the toddler height dataset, the K-Nearest Neighbors (KNN) algorithm proved to be the model with the best performance, with an accuracy of 99.91%. This value exceeded that of the Decision Tree, which achieved an accuracy of 99.89%, and the SVM (RBF) algorithm, which achieved 98.48%.]]>
Copyrights © 2025
