<![CDATA[Stunting is a chronic nutritional problem affecting child growth, particularly in regions with high prevalence, such as Sulawesi Island. This study aims to compare two optimization methods, Artificial Bee Colony (ABC) and Gravitational Search Algorithm (GSA), in the Fuzzy Geographically Weighted Clustering (FGWC) analysis to group regencies and cities based on factors contributing to stunting. The data used included health and socio-economic indicators from 66 regencies/cities in Sulawesi Island. Three validity indices—Classification Entropy (CE), Separation Index (SI), and Xie and Beni’s Index (XB)—were employed to assess clustering performance. The findings indicate that the FGWC-ABC method outperformed FGWC-GSA, yielding lower CE and XB values and a higher SI value, signifying better clustering results. The FGWC-ABC method, at a fuzziness value of m = 1.5, formed two clusters: Cluster 1, comprising 49 regencies/cities with relatively lower stunting prevalence and better socio-economic conditions, and Cluster 2, consisting of 17 regencies/cities with higher stunting prevalence and poorer socio-economic conditions. This study highlights the potential of FGWC-ABC in optimizing regional classification for targeted interventions in addressing stunting. The results provide a valuable reference for policymakers in designing effective strategies to mitigate stunting issues.]]>
Copyrights © 2025
