CLUSTERING DAERAH RAWAN STUNTING DI INDONESIA MENGGUNAKAN ALGORITMA K-MEANS

Stunting is one of the main nutritional problems in Indonesia whose prevalence is still high, which is 21.5% in 2023, exceeding the WHO standard which should be below 20%. This study aims to identify areas at risk of stunting in Indonesia using the K-Means method based on factors that cause stunting. The data used includes coverage of essential services and stunting cases from 38 provinces during 2022-2024, obtained from the Ditjen Bina Bangda Kemendagri dashboard as much as 3534 data. The research begins by determining the variables that have a stronger correlation with stunting among other variables in each year using Spearman correlation. Then, the optimal number of clusters was determined using the Silhouette Coefficient method. The results show that the optimal number of clusters for each year is K=2 with each value being 0.387 (2022), 0.436 (2023), and 0.446 (2024). Cluster 1 represents high stunting risk areas with low coverage of essential services, while cluster 2 represents low stunting risk areas with more optimized essential services. In 2022, cluster 1 consists of 25 provinces, and cluster 2 consists of 13 provinces. In 2023, cluster 1 decreases to 18 provinces and cluster 2 increases to 20 provinces. In 2024, cluster 1 declines further to 14 provinces, while cluster 2 increases to 24 provinces.