A Comparative Study of K-Means and K-Medoids for Clustering Dengue Fever Risk Areas in Medan

Dengue Hemorrhagic Fever (DHF) is a localized disease that continues to contribute to a high number of cases in Medan City. The local health authority faces challenges in identifying priority areas for effective prevention and control. This study applies data clustering techniques to map DHF risk areas by comparing the performance of K-Means and K-Medoids algorithms. The optimal number of clusters was determined using the Silhouette Coefficient, while the clustering quality was assessed using the Davies-Bouldin Index (DBI). The findings indicate that K-Means performs best with four clusters and achieves a lower DBI value compared to K-Medoids. Based on this, the study recommends using K-Means to categorize DHF risk areas into four priority levels: high, medium, low, and very low. This approach is expected to support the Medan City Health Office in implementing more targeted and efficient DHF control strategies.