<![CDATA[Poverty is one of the main challenges faced by the government in its efforts to improve people's welfare. Identifying regions based on the poverty line level is an important step to ensure well-targeted interventions. This study aims to categorize districts/cities based on poverty levels using the K-Means Algorithm, so that it can be a guide in data-based policy making. The research method starts with data collection, data selection process to handle missing values using the replacement method. Determination of the optimal number of clusters was done using Within Sum of Squares (WSS) to ensure that each region was grouped into clusters based on their level of similarity, which showed that three clusters were the ideal number. An evaluation of the clustering results was conducted to ensure the stability and accuracy of the clustering. The results show that the districts/municipalities are divided into three clusters based on the poverty line level, namely cluster 0 with a high poverty line level (241 regions), cluster 1 with a medium poverty line level (247 regions), and cluster 2 with a low poverty line level (90 regions). This study concludes that the K-Means Algorithm is effective in clustering regions based on poverty levels, providing a strong basis for data-driven decision-making. Future research is recommended to use more diverse data and cover more indicators, such as education level, access to health services, or infrastructure quality.]]>
Copyrights © 2025
