K-Means Binary Search Centroid With Dynamic Cluster for Java Island Health Clustering

This study is focused on determining the health status of each district/city in Java using K-means Binary Search Centroid and Dynamic Kmeans algorithms. The research data uses profile Island 2020. Comparative algorithms were tested Davies Bound Index Calinski-Harabasz methods traditional k-means algorithm dynamic binary search centroid k-means. Based test, 5 clusters found distribution area, including 11 regions with very high quality cluster 1, 24 quality, 28 moderate 4 low 45 regions, deficient best validation value DBI 1.8175 CHI 67.7868. Overall optimization based results a better average smaller number iterations than algorithm. test can be used as one evaluating level area reference for decision-making policies related agencies