<![CDATA[The terms "underdeveloped" and "non-underdeveloped" regions highlight the gap between regions in Indonesia. The government determines the status of underdeveloped regions every five years. Presidential Decree No. 63 of 2020 determines 62 districts/cities in Indonesia as underdeveloped regions. This study aims to classify disadvantaged district status using the Support Vector Machine (SVM) algorithm with three kernel types: linear, polynomial, and Radial Basis Function (RBF). SVM was selected for its effectiveness in handling high-dimensional data and non-linear classification tasks. The dataset, sourced from BPS and JDIH in 2022, comprises 20 variables covering socioeconomic, infrastructure, and public service indicators. The data distribution is imbalanced, with only 62 out of 514 districts labeled as disadvantaged. Optimal parameters were determined experimentally: linear (C = 0.1), polynomial (C = 1, d = 3), and RBF (C = 1, γ = 0.1). Based on evaluation results, the linear kernel achieved the best performance on the given dataset, with an accuracy of 0.94, precision of 0.91, recall of 0.81, and F1-score of 0.85. The model classified 45 districts as disadvantaged and 469 as non-disadvantaged. A total of 29 districts showed discrepancies compared to the official classification. These differences may indicate either changing ground conditions or limitations in policy criteria, highlighting the potential of data-driven approaches to support more targeted and equitable regional development planning.]]>
Copyrights © 2025
