Penggerombolan Desa di Jawa Barat Berdasarkan Daerah Rawan Bencana
DOI:
https://doi.org/10.21009/JSA.06210Keywords:
clustering, K-Prototypes, silhouette coefficientAbstract
Indonesia is one of the countries that has a large potential for natural disasters. Indonesia's position at the confluence of 4 continental plates makes the potential for earthquakes even greater. The tropical climate with 2 seasons makes changes in weather, temperature and wind direction quite extreme. These climatic conditions combined with the relatively diverse surface and rock topography conditions, these conditions can cause several bad consequences for the community such as hydrometeorological disasters such as floods, landslides, forest fires, and droughts. Particularly in West Java province, natural disasters that have occurred include: landslides, droughts, cyclones/typhoons, tidal waves, fires, volcanic eruptions, tsunamis, and other disasters. The purpose of this study was to cluster villages in the West Java region based on the level of disaster-prone in 2018. The research was carried out using K-Prototypes clustering and testing evaluation using the silhouette coefficient. The results showed that the optimal number of clusters in this study was nine clusters. These clusters can be distinguished based on the disaster category and the characteristics of the area.