<![CDATA[Abstract — The Family Hope Program (PKH) is a conditional social assistance scheme aimed at poor and vulnerable households. However, the process of determining recipients at the village level is potentially subjective because it is carried out through manual assessments of various household socioeconomic characteristics. This study aims to classify recipient eligibility using the Decision Tree algorithm based on secondary village administrative data from 2025 for 493 households. The variables used include the number of family members, the head of the family’s occupation, family income, house condition, and aid receipt status as target variables with two classes: receiving and not receiving. The analysis stages follow the Knowledge Discovery in Database framework, including data selection, preprocessing, transformation, data mining, and evaluation. Preprocessing is carried out by checking for duplicate data, empty data, and category format errors, while transformation is carried out by grouping numeric values into specific categories. Modeling is carried out using RapidMiner with the Decision Tree algorithm and model evaluation using a confusion matrix. The results show that house condition is the most dominant attribute, followed by family income. The model produces an accuracy of 80.61%, a precision for the receiving class of 87.50%, and a recall of 65.12%. These findings indicate that Decision Trees can assist the process of identifying aid recipients in a more structured, consistent, and data-based manner, but still need to be validated through field verification. Key word — confusion matrix; data mining; decision tree; family hope program; social aid]]>
Copyrights © 2026
