<![CDATA[Dengue Hemorrhagic Fever (DHF) remains a major public health problem in Lamongan Regency, Indonesia, with unequal distribution across sub-districts. This study aims to identify and classify DHF-prone areas using Spatial ‘K’luster Analysis by Tree Edge Removal (SKATER), a graph-based spatial clustering method. The study used cross-sectional secondary data at the sub-district level, including DHF Incidence Rate (IR), population density, rainfall, and percentage of adequate sanitation. Spatial autocorrelation was analyzed using Moran’s Index, followed by weighted graph construction and Minimum Spanning Tree (MST) partitioning for cluster formation. Cluster quality was evaluated using the Sum of Squared Deviations (SSD) and Between-Cluster Sum of Squares (BSS). The Moran’s I results showed significant spatial autocorrelation for all variables (p < 0.05). The five-cluster configuration produced better clustering performance, with lower SSD (49.84) and higher BSS (58.16) compared to the three-cluster configuration (SSD = 86.18; BSS = 21.82). The results revealed spatial variations in DHF vulnerability, ranging from very low to very high categories. These findings indicate that the SKATER method effectively identifies geographically contiguous and homogeneous DHF-prone areas to support spatially targeted DHF control planning in Lamongan Regency]]>
Copyrights © 2026
