The number of cases of the spread of dengue hemorrhagic fever (DHF) in Beringin Country Village every year has increased for this reason, it is necessary to handle dengue hemorrhagic fever so that the spread does not increase dengue hemorrhagic fever cases every year. The vast natural conditions of Banyan Country Village make it difficult to monitor the spread of diseases based on the most cases. This causes slow prevention and control of the disease. For this reason, it is necessary to group or cluster diseases based on the number of cases that occur each year. This study aims to apply the K-Medoids algorithm to group dengue hemorrhagic fever data based on the number of cases each year. The process of solving problems can be done by processing data. Data mining is a data processing process for extracting information stored in the data set, in data mining the process of grouping data for the process of handling dengue hemorrhagic fever is included in clustering techniques. The K-medoids algorithm is a limiting clustering method for grouping a collection of objects into clusters. The data used is in the form of data on dengue sufferers, from 2017-2022. Test results were obtained for high cluster dengue disease of 72 members and low 28 members. The K-medoids algorithm can help facilitate the process of forming dengue hemorrhagic fever treatment clusters. Testing using manual calculations and the Rapidminer application gets the same results as the system. This shows that the system has worked well.