Poverty is a complex problem that hampers socio-economic development in Indonesia, especially in Central Java Province, which encounters significant challenges, with a poverty rate reaching 10.77% in 2023. This study aims to identify spatial patterns of poverty in 35 districts/cities in Central Java Province by grouping areas based on the number of poor individuals reported by the Central Java Province Statistics Agency (BPS) in 2023. The Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm groups districts/cities based on poverty data density with optimized parameters to produce statistically significant clusters. The results of the analysis reveal four clusters, specifically cluster 0 (moderate poverty), cluster 1 (high poverty), cluster 2 (very high poverty), and cluster 3 (low poverty). Model validation was executed using the Silhouette Score (0.447) and Davies-Bouldin Index (0.441), which showed the validity of the clustering. This study is anticipated to provide strategic implications for the Central Java Provincial Government in formulating more effective poverty alleviation policies, such as resource allocation adjusted to each cluster's characteristics. In addition, this study enables future exploration of additional socio-economic factors influencing poverty, such as the Human Development Index, education, health, infrastructure, resource accessibility, and comparative analysis of clustering algorithms for enhanced accuracy.