Analisis Pola Penyakit Kronis pada Lansia Menggunakan K-Means Clustering di Puskesmas Kelurahan Semper Barat

Abstract

Chronic diseases have become a major health problem as well as the leading cause of death worldwide. The main chronic diseases causing death globally are cardiovascular diseases, chronic respiratory diseases, and metabolic diseases such as diabetes (WHO, 2018). Semper Barat Subdistrict is an area with a significant number of elderly residents, with healthcare services centered at the Semper Barat Community Health Center (Puskesmas). However, so far, there has been no study that specifically analyzes the patterns of chronic diseases among the elderly in this area using a data mining approach. This study presents a novelty in the form of a case study on clustering chronic diseases within the community of the subdistrict using data mining with the K-Means algorithm. The results show that this model is capable of providing precise values in clustering chronic diseases. The clustering results can be utilized by the Semper Barat Community Health Center as a basis for decision-making in conducting targeted outreach and treatment, thereby facilitating better access to elderly individuals who already have a history of chronic diseases according to their disease group. The testing results from the previous Cluster Distance Performance showed an evaluation value of 0.579 for two clusters, which was the closest to zero compared to other numbers of clusters. In the context of the K-Means algorithm, values closer to zero indicate that the data within a cluster have greater similarity, and the distance between clusters is sufficiently distinct.