<![CDATA[Jakarta, as a metropolitan city in Indonesia, often experiences flooding caused by high rainfall, poor drainage systems, and rapid urbanization. This research aims to classify flood-prone areas in Jakarta using a combination of K-Means Clustering and Naïve Bayes Classifier algorithms. The research phase begins with data collection from the Satu Data Jakarta website, including attributes such as region, sub-district, village, average water level, number of affected RWs, number of affected families, number of affected people, and number of flood events. The collected data is then processed through cleaning and normalization stages before being analyzed using the K-Means algorithm to group areas based on their flooding characteristics. Furthermore, the Naïve Bayes algorithm was used to build a classification model that predicts flood-prone areas. The results showed that the combination of these two algorithms resulted in higher average accuracy compared to the use of conventional Naïve Bayes, having an accuracy of 98.18%% at training and testing data split ratios of 70:30, 80;20 and 90:10. The findings provide valuable insights for flood risk mitigation in Jakarta, assisting the government in taking more effective preventive measures.]]>
Copyrights © 2025
