<![CDATA[Purpose: Malnutrition in toddlers is a nutritional issue that Indonesia is still dealing with. Toddlers can suffer from decreasing cognitive and physical abilities, as well as being categorized as having a high risk of death. Early detection is crucial for preventing this and providing appropriate treatment if malnutrition is detected. Classification is a machine-learning technique widely used in disease detection. Because it is simple and easy to implement, K-Nearest Neighbor is the most used classification algorithm. Detecting malnutrition can be done automatically and more quickly by utilizing classification and machine learning algorithms. The aim of this study was to analyze performance to find out which model is best for detecting malnutrition by evaluating the performance of classification using KNN with the Euclidean distance function.Methods: The dataset used in this study is the nutritional status of toddlers from Puskesmas Ubung. The classification method proposed in this research is the KNN algorithm with Euclidean distance. There are three scenarios for the classification model that will be used. Performance classification will compare each model in terms of accuracy, precision, recall, f1-score, and mean absolute error.Results: The experimental results show that KNN k = 15 using the first model generates excellent classification when classifying malnourished toddlers using the Euclidean distance function. The model obtains 91% accuracy, 86.6% precision, 83.8% recall, 85.2% recall, and a mean absolute error of 0.09.Novelty: In this experiment, we analyzed the performance of the KNN to classify malnourished children using a nutritional status dataset, which resulted in an excellent classification that could be used for early detection.]]>
Copyrights © 2023
