Clustering is a process of grouping data sets into several groups. K-Mean is an innovation or new knowledge for puskesmas in calculating patient disease data. This study aims to determine the clustering of medical record data using the k-mean algorithm. This type of research uses descriptive quantitative data. The population of medical record data taken is the last 3 months of 2020. The type of data used is secondary data, data collection by observation and data analysis using thealgorithm K-Mean. The result of the research is that there are 3 clusters determined. Among them are the Low Suffering Disease Cluster, Medium Suffered Disease and High Suffered Disease Cluster. Low suffering disease which is coded to cluster A there are 97 patients with a percentage (14%), high suffering disease is coded to cluster B there are 318 patients with a percentage (45%), and moderate illness is coded to cluster C there are 292 patients with a percentage (41%). Disease The most common at the Lubuk Alung Health Center are Diabetes, Schizophrenia, Hypertension, Stroke, Refractive Disorders, Rheumatism, Tinea, Gout, Epilepsy and Cataracts. From this pattern, the researcher compared with several sources, that the pattern of the disease is indeed suffered by a lot of people in the age range above 30 years. The K-Mean method can be a new innovation that is expected to make it easier to calculate patient disease data at the Lubuk Alung Health Center.
                        
                        
                        
                        
                            
                                Copyrights © 2021