Development of a Web-Based Application for Verifying the Authenticity of Digital Population Documents Using Extreme Gradient Boosting (XGBoost)

Alvian Bastian; Andi Arma Gansa; Muh. Irsan S

doi:10.31963/elekterika.v23i1.5775

Authors

Alvian Bastian Politeknik Negeri Ujung Pandang
Andi Arma Gansa Politeknik Negeri Ujung Pandang
Muh. Irsan S Politeknik Negeri Ujung Pandang

DOI:

https://doi.org/10.31963/elekterika.v23i1.5775

Keywords:

XGBoost, Digital Population Documents, Document Verification, Image Forensics, ELA-CNN

Abstract

Along with the increasing use of digital population documents in Indonesia, the risk of document forgery has become a serious challenge that threatens data validity in various administrative processes. This research aims to develop and evaluate the performance of a website-based application capable of automatically verifying the authenticity of digital population documents. The proposed method integrates the Image Forensics approach with Machine Learning.The system extracts a series of features from the document image , including metadata analysis , Error Level Analysis (ELA) processed using a Convolutional Neural Network (CNN) , and image statistical features. This feature set is then analyzed using the Extreme Gradient Boosting (XGBoost) algorithm to classify the document as "Authentic" or "Fake".The application is built with an e-Government architecture, using Laravel for the frontend and Flask as the API server for the analysis process. Performance testing results using a confusion matrix against 90 data samples show that the system successfully achieved an accuracy level of 93.33%. The average response time for one complete prediction cycle was recorded at 12.66 seconds. These results prove that the developed system is an effective and efficient solution for mitigating the risk of digital population document forgery.

References

[1] Kementerian Kominfo/Direktorat Jenderal Aplikasi Informatika, Satu dekade pembangunan digital Indonesia 2014–2024. Jakarta, 2024. Accessed: Sep. 09, 2025. [Online]. Available: https://aptika.kominfo.go.id

[2] A. Fachmi, “PENINGKATAN KOMPETENSI PENGELOLA ARSIP DENGAN KEAHLIAN DOKUMEN FORENSIK,” Info Bibliotheca: Jurnal Perpustakaan dan Ilmu Informasi, vol. 5, no. 1, pp. 72–88, Dec. 2023, doi: 10.24036/ib.v5i1.447.

[3] W. , Anggara, P. , Hafidati, and M. , Kamil, “PERTANGGUNGJAWABAN TINDAK PIDANA PEMALSUAN SURAT KTP YANG DAPAT MENGAKIBATKAN KERUGIAN PADA ORANG LAIN,” JURNAL PEMANDHU, vol. 5, no. 1, pp. 229–250, 2024.

[4] K. E. Purnama, C. Rozikin, and A. A. Ridha, “ANALISIS FORENSIC CITRA DIGITAL MENGGUNAKAN TEKNIK ERROR LEVEL ANALYSIS DAN METADATA BERDASARKAN METODE NIST,” 2023.

[5] R. Abrori, M. Fitria, and H. Bullah, “Mendeteksi Orisinalitas Citra Menggunakan Teknik Error Level Analysis dan Metadata,” Indo-Fintech Intellectuals: Journal of Economics and Business, vol. 4, no. 2, pp. 170–178, Jun. 2024, doi: 10.54373/ifijeb.v4i2.1169.

[6] W. D. Prastowo, D. Danianti, and A. Pramuntadi, “ANALISIS RISIKO PADA PENGEMBANGAN PERANGKAT LUNAK MENGGUNAKAN METODE AGILE DAN RAD (RAPID APPLICATION DEVELOPMENT),” Citizen : Jurnal Ilmiah Multidisiplin Indonesia, vol. 3, no. 3, pp. 169–174, Aug. 2023, doi: 10.53866/jimi.v3i3.388.

[7] M. F. Rahman, R. Rachman, J. Putra, and Y. Wihardi, “Analisis Statistik dan Implementasi Image Masking Berdasarkan Hasil Error Level Analysis Pada Gambar Digital Statistical Analysis and Image Masking Implementation Based on the Results of Error Level Analysis on Digital Images,” 2020. [Online]. Available: https://ejournal.upi.edu/index.php/JATIKOM

[8] F. Harahap, “Deteksi Foto Manipulasi Dengan Tools Forensicallybeta dan Imageforensic.org Dengan Metode Error Level Analysis (ELA),” 2021.

[9] M. Muchtar and R. A. Muchtar, “PERBANDINGAN METODE KNN DAN SVM DALAM KLASIFIKASI KEMATANGAN BUAH MANGGA BERDASARKAN CITRA HSV DAN FITUR STATISTIK,” Jurnal Informatika dan Teknik Elektro Terapan, vol. 12, no. 2, Apr. 2024, doi: 10.23960/jitet.v12i2.4010.

[10] D. Nasution, “Prosiding SNASTIKOM: Seminar Nasional Teknologi Informasi & Komunikasi Paper Klasifikasi Objek Menggunakan Metode Convolutional Neural Network (CNN),” 2022.

[11] S. Yuliany and A. Nur Rachman, “Implementasi Deep Learning pada Sistem Klasifikasi Hama Tanaman Padi Menggunakan Metode Convolutional Neural Network (CNN),” 2022.

[12] E. H. Yulianti, O. Soesanto, and Y. Sukmawaty, “Penerapan Metode Extreme Gradient Boosting (XGBOOST) pada Klasifikasi Nasabah Kartu Kredit,” JOMTA Journal of Mathematics: Theory and Applications, vol. 4, no. 1, 2022.