<![CDATA[Malnutrition is one of the foremost health problems experienced by children under five in many countries, especially in low and middle-income countries. Meanwhile, the target of Sustainable Development Goals (SDGs) 2.2 is that all forms of malnutrition must end by 2025. Therefore, this study aims to predict the toddlers’ nutritional status (malnutrition, undernutrition, overnutrition, and normal nutrition) based on age, body mass index (BMI), weight, and length using the Multinomial Logistic Regression (MLR) classification method. The dataset consists of two hundred toddlers obtained from the Kaggle site. Following pre-processing, the dataset is divided, with 80 percent of the data for training and the remaining 20 percent for testing. The model was trained using 10-fold cross-validation (CV). In Addition, the MLR model performance was evaluated using the confusion matrix (CM), the area under the curve (AUC), and the Kappa coefficient (KC). The evaluation results using CM show that the accuracy, sensitivity, and specificity values are 0.9412, 0.9375, and 0.9790, respectively. AUC and KC also show excellent results. It indicates that the MLR method is an esteemed and recommended method for predicting the nutritional status of toddlers. Therefore, this research can contribute to providing early information so that the Government can immediately determine the necessary treatment.]]>
Copyrights © 2024
