Classification of Malnutrition Risk in Toddlers Using the C4.5 Method (Case Study: UPTD Gunung Meriah Health Center)

Abstract

Abstract :The problem of malnutrition in toddlers requires early identification through systematic anthropometric data processing. This study aims to apply the C4.5 algorithm to classify the nutritional status of toddlers, measure model accuracy, and identify influential attributes at the UPTD Gunung Meriah Health Center. This study uses a quantitative approach with a data mining classification method based on Decision Tree C4.5. The research data consists of documentation of 386 toddlers, consisting of 343 data with good nutrition, 26 data at risk of malnutrition, and 17 data with severe malnutrition. The research instruments are toddler datasets and data processing software. Predictor attributes include age, gender, weight, height, and mid-upper arm circumference (MUAC), while nutritional status is the target variable. The analysis was carried out through preprocessing, decision tree formation, functional testing Black Box Testing, and confusion matrix evaluation with a division of training data and test data of 80:20. The results showed a classification accuracy of 87.18%, with MUAC as the main attribute and weight as a supporting attribute. All main functions of the web-based system run according to the test scenario. In conclusion, the C4.5 algorithm has the potential to support early identification of toddler nutritional status and systematic data management, but the classification results still require verification by health workers.