Analisis Faktor Risiko Stunting pada Balita di Desa Kesetnana Menggunakan Metode Random Forest

Abstract

Stunting is a growth disorder triggered by chronic malnutrition, impacting the physical and cognitive development of young children. Kesetnana Village in South Central Timor Regency records a high prevalence of stunting. This study aims to classify stunting status using the Random Forest Classifier algorithm and assess its performance. The quantitative analysis was conducted on secondary data from 1,451 toddlers obtained through total sampling from the Health Center and Kesetnana Village Office in 2023. The variables analyzed include birth weight and height, measurement age, as well as Z-scores for Height/Age, Weight/Age, and Weight/Height. Data were processed using Python on the Google Colaboratory platform, with 75% allocated for training and 25% for testing. The findings indicate that birth weight, measurement age, and height are the primary factors in stunting classification. The model achieved 97% accuracy, with high precision and recall values, demonstrating its effectiveness in classifying stunting. This model can be utilized by health professionals and policymakers to identify stunting risk at an early stage and design targeted nutritional interventions in high-prevalence areas.