<![CDATA[General Background: Voter participation serves as a core indicator of democratic quality and civic awareness. Specific Background: In East Java’s Mataraman region, significant disparities in electoral participation highlight socioeconomic influences measurable through the Village Development Index (IDM). Knowledge Gap: No prior research integrates IDM-based indicators with machine learning methods for voter behavior prediction. Aims: This study develops a classification model using C4.5, Naïve Bayes, and SVM algorithms to predict voter participation based on IDM attributes. Results: The Decision Tree C4.5 algorithm achieved the highest accuracy (80.87%) and F1-score (0.88) compared to Naïve Bayes and SVM, identifying education and healthcare access as primary determinants of high participation. Novelty: The integration of IDM and C4.5 classification introduces a novel framework for data-driven political participation analysis. Implications: The model can assist policymakers and electoral bodies in targeting civic engagement initiatives within underrepresented regions. Highlights: C4.5 algorithm effectively predicts voter engagement. Education and health access influence participation. Data-driven policy enhances democratic quality.]]>
Copyrights © 2025
