<![CDATA[This study aims to analyze and cluster poverty data in Central Java using the Self-Organizing Maps (SOM) method, an approach in unsupervised learning that is efficient in mapping multidimensional data into two-dimensional representations. The poverty data used includes various socio-economic indicators, such as income, education, health access, and housing conditions. By applying SOM, this study attempts to identify hidden patterns and relationships between variables that contribute to poverty in each region in Central Java. The results of this clustering are expected to provide deeper insight into the characteristics and distribution of poverty, as well as assist in making more targeted policies in poverty alleviation efforts. This study shows that the SOM method is able to effectively group areas with similar poverty characteristics, and provide visualizations that facilitate understanding of the complexity of poverty data in Central Java. The implementation of this method is able to produce 4 groups / clusters of poverty levels which are expected to be the basis for further research in socio-economic mapping, as well as a tool in planning and evaluating poverty alleviation programs at the regional level.]]>
Copyrights © 2025
