Unlocking potential of data: A localized data-driven approach for stunting reduction in South Kalimantan Province

Abstract

This study addresses the issue of stunting in South Kalimantan Province, where high stunting prevalence rates persist. Through a comprehensive analysis of factors influencing stunting prevalence, predictive modeling using machine learning, and clustering analysis of districts based on stunting rates, the research aims to support the provincial government in formulating effective and sustainable strategies. The findings highlight influential factors such as HDI, poverty rates, immunization coverage, breasfed babies, number of uninhabitable houses, and access to clean water. The study also utilise machine learning to build model that aids in predicting future stunting prevalence, while clustering analysis categorizes districts into distinct groups. These insights guide the government in prioritizing interventions, setting prevalence targets, and determining strategic areas for stunting reduction efforts.