<![CDATA[Floods are among the most frequent natural disasters occurring in West Java Province and have significant impacts on social and economic conditions. Although the government provides flood incident frequency data down to the village and sub-district levels, its utilization for detailed vulnerability analysis remains limited. This study aims to classify regions based on the frequency of flood events using the K-Means Clustering method as an analytical approach to produce a more comprehensive risk mapping. The dataset consists of flood incident records from 2022 to 2024 obtained from official government sources. The analytical process follows the stages of Knowledge Discovery in Database with a primary focus on the implementation of the K-Means algorithm, while model evaluation is conducted using the Elbow Method and Silhouette Score to determine the optimal number of clusters. The results indicate that three clusters provide the most structured grouping of flood risk. The low-risk cluster consists of 12,454 regions that experienced zero flood events. The medium-risk cluster includes 3,404 regions with flood frequencies ranging from 1 to 6 events. Meanwhile, the high-risk cluster comprises 76 regions with flood occurrences between 7 and 33 events. These findings are expected to support flood mitigation planning, spatial planning strategies, the development of flood-control infrastructure, and to assist communities in evaluating the safety of potential residential areas.]]>
Copyrights © 2025
