Dengue hemorrhagic fever (DHF) is an infectious disease caused by the dengue virus. This disease mostly affects children and adults, and also the number of sufferers and the extent of this disease are increasing. In grouping patient data based on the parameters of the symptoms of dengue fever, we will use the clustering method. In this study, a comparison of two clustering algorithms, namely K-Means and K-Medoids, uses a distance measure, namely Manhattan Distance, in both algorithms to determine the most optimal algorithm for grouping dengue patient data. The results of clustering carried out with a total of 3 k in the K-means clustering algorithm have members, namely in cluster 0 there are 32 members, cluster 1 there are 15 members, and cluster 2 there are 26 members. While the K-Medoids clustering algorithm has members, namely in cluster 0 there are 46 members, cluster 1 there are 24 members, and cluster 2 there are 3 members. Based on the results of the cluster obtained, the K-Medoids algorithm has a placement error in the data to 27, 34, 48, and 70 to the right cluster. While in the K-Means algorithm there is no error in the placement of data in the cluster. The results of the cluster analysis performed based on the Silhouette Coefficient value on the dengue patient data resulted that the K-Means algorithm was the optimal clustering method with the highest Silhouette Coefficient value of 0.25362003217634904 compared to the K-Medoids method of 0.1821645521976126. So, it can be concluded that the clustering method with the K-Means algorithm has a good cluster quality compared to the K-Medoids algorithm.