<![CDATA[The level of poverty serves as a significant indicator influencing a nation's well-being. Poverty can arise from various factors, such as limited job opportunities resulting in insufficient income to cover living expenses, substantial family responsibilities, and more. In this context, the government plays a role by providing assistance, such as social aid programs. One step in providing this assistance involves individuals being registered as participants in the Unified Social Welfare Data (Data Terpadu Kesejahteraan Sosial or DTKS). Becoming a DTKS participant requires meeting the criteria categorizing someone as extremely poor, which is generally determined by the Minister of Social Affairs' Decision No. 146/HUK/2013 on the criteria for registered individuals in extreme poverty.This consideration can serve as a guideline in determining the socioeconomic status of community groups within a region. However, on a larger scale, classifying communities based on poverty levels can be a complex and time-consuming task. K-Means clustering is one of several non-hierarchical data clustering methods that work by partitioning existing data into one or more clusters or groups. This clustering can be applied to categorize a large dataset to enhance the accuracy of the information obtained, such as assessing the poverty level in a specific area. The objective of this research is to develop an application that facilitates the categorization of communities in analyzing the progression of poverty rates in a region based on predefined criteria. This aids the government and other stakeholders in understanding poverty distribution better, identifying high-risk groups, and designing targeted and effective social aid programs or policies. The outcomes of this research showcase visualizations depicting the percentage composition of each group within a dataset. The presented data visualizations can also be customized based on categories such as the number of clusters, regions, years, and more.]]>
Copyrights © 2024
