PENERAPAN DATA MINING UNTUK CLUSTERING POTENSI KESEHATAN MASYARAKAT DESA MENGGUNAKAN ALGORITMA K-MEANS

Abstract

The effective use of health data is essential to support planning and decision-making in village-level community health programs. This study aims to classify the health potential of the Cigugur Village community based on the risk level of infectious diseases using a data mining approach. The research adopts a quantitative descriptive method and follows the Knowledge Discovery in Database (KDD) process, including data collection, preprocessing, clustering, and visualization. Data were obtained from the Cigugur Community Health Center (Puskesmas) for the period 2023–2024, consisting of patient visit records, residential hamlets, and types of infectious diseases. Data analysis was conducted using the K-Means clustering algorithm, with the optimal number of clusters determined through the Elbow Method. The results show three clusters representing high-, medium-, and low-risk zones. Palumbungan and Cipaku were classified as high-risk areas, Tegalega as medium-risk, and Cilembu and Karang Anyar as low-risk. The findings indicate that K-Means clustering is effective in mapping community health conditions and can support data-driven decision-making for prioritizing village health programs.