Klasifikasi Data Balita Stunting dengan Metode Decision Tree

Abstract

Stunting is a chronic malnutrition condition that significantly impacts the physical growth and cognitive development of toddlers, often leading to irreversible damage to physical and mental capabilities during the critical first 1,000 days of life. In Indonesia, stunting remains a critical public health concern that requires early and accurate detection to mitigate long-term adverse effects. Conventional methods of determining nutritional status often rely on manual measurements and look-up tables, which can be time-consuming and susceptible to human error when processing large datasets. This study aims to address these challenges by developing an automated classification model for toddler stunting status using the Decision Tree algorithm. The research methodology includes data collection, preprocessing (cleaning and attribute selection), and model training using the Python programming language within the Google Colab environment, leveraging the Scikit-Learn library for efficient computation. The dataset utilized comprises key anthropometric attributes such as age, gender, body height, and weight. The experimental results demonstrate that the Decision Tree model effectively classifies nutritional status with an accuracy of 99,91%, indicating a high degree of reliability for practical implementation. Furthermore, the model generates interpretable decision rules, enabling healthcare practitioners to easily understand the primary determinants of stunting. Consequently, this model reliably aids in early stunting detection and prevention, potentially facilitating real- time monitoring in remote areas where access to specialized pediatric care is limited.