PENGELOMPOKAN KECAMATAN DI JAKARTA BERDASARKAN TINGKAT KERAWANANNYA TERHADAP BANJIR MENGGUNAKAN K-MEANS CLUSTERING

The purpose of this study was to cluster the sub-districts in Jakarta based on their level of vulnerability to flooding. This study uses a quantitative method using k-means clustering. The number of clusters in this study amounted to 3 clusters, namely districts with a high level of vulnerability, districts with moderate vulnerability and districts with low vulnerability. There are 9 variables used in this study, namely the number of neighbourhood and hamlet that have an impact on flooding, the minimum water level during the flood, the maximum water level during the flood, the length of inundation during the flood, the number of refugee camps for flood victims, the number of refugees during the flood, number of victims who died during the flood and the number of camps for refugees. After grouping, it was found that there were 8 districts that had a high level of vulnerability, 29 districts with a moderate level of vulnerability, and 6 districts with a low level of vulnerability. Keyword: vulnerability, flood, sub-district, cluster, k-means clusterin