A Simple Machine Learning Application for Deepfake Detection and Analysis of How It Works

Ilham Syahnara

doi:10.65475/t37afz44

Authors

Ilham Syahnara Universitas Muhammadiyah Ponorogo Author

DOI:

https://doi.org/10.65475/t37afz44

Keywords:

Deepfake Detection, MesoNet-4, Convolutional Neural Network (CNN), Domain Shift, Cross-Domain Generalization

Abstract

The rapid advancement of deepfake technology poses a serious threat to digital security. Unfortunately, the majority of existing detection systems still rely on complex network architectures with heavy and costly computational requirements. Therefore, this study evaluates the reliability of MesoNet-4 a lighter Convolutional Neural Network (CNN) architecture and efficiently detecting deepfakes. The evaluation was conducted using a dataset of 140,000 faces from Kaggle, simulated in the Google Colab environment. Experimental results show that the model achieved a high accuracy rate of 94.00% when tested on data from the same distribution (in-dataset). However, this performance dropped significantly to 50.00% when subjected to out-of-dataset testing. In this scenario, the model experienced a complete detection failure by classifying all fake image samples as real images. This phenomenon indicates that lightweight architectures are highly susceptible to overfitting to the specific photographic characteristics of the training data, thereby failing to adapt to domain shifts in real-world conditions. This study concludes that high accuracy in a controlled environment does not guarantee the model’s practical reliability. Moving forward, the development of detection systems needs to prioritize cross-domain generalization capabilities and integrate dynamic preprocessing modules, such as automatic face detection.

Downloads

Download data is not yet available.

References

[1] M. A. Amin, Y. Hu, C.-T. Li, and B. Liu, "Deepfake detection based on cross-domain local characteristic analysis with multi-domain transformer," Alexandria Engineering Journal, vol. 91, pp. 592–609, 2024.

[2] J. Li, H. Xie, J. Li, Z. Wang, and Y. Zhang, "Frequency-aware Discriminative Feature Learning Supervised by Single-Center Loss for Face Forgery Detection," arXiv preprint arXiv:2103.09096, 2021.

[3] A. V. Nadimpalli and A. Rattani, "On Improving Cross-dataset Generalization of Deepfake Detectors," dalam CVPR Workshops, 2022.

[4] K. Shiohara and T. Yamasaki, "Detecting Deepfakes with Self-Blended Images," dalam Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022.

[5] H. Zhao, W. Zhou, D. Chen, T. Wei, W. Zhang, and N. Yu, "Multi-attentional Deepfake Detection," dalam Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021.

[6] L. Lin, X. He, Y. Ju, X. Wang, F. Ding, and S. Hu, "Preserving Fairness Generalization in Deepfake Detection," arXiv preprint arXiv:2402.17229, 2024.

[7] L. Chen, Y. Zhang, Y. Song, L. Liu, and J. Wang, "Self-supervised Learning of Adversarial Example: Towards Good Generalizations for Deepfake Detection," dalam Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022.

[8] R. E. Ardiansyah and J. Karaman, "A Decision Support System For Roof Tile Selection Using AHP In The Bedingin Roof Tile Industrial Center," Journal Information System and Computer Application, PT Mekar Research and Publishing, Apr. 2026, doi: 10.65475/vde80844.

[9] S. A. Gulo, A. A. Pertiwi, S. P. S. Nasution, and H. Syahputra, "Deteksi Deepfake Dalam Citra Menggunakan Convolutional Neural Network (CNN)," JATI (Jurnal Mahasiswa Teknik Informatika), vol. 9, no. 5, pp. 8655–8660, Oct. 2025.

[10] Irfan and R. Arijanto, "Penggunaan Machine Learning Untuk Deteksi Kerusakan Pada Security Seal Melalui Citra Digital Menggunakan Algoritma Convolutional Neural Network," POTERS: Proceedings of Technology, Engineering and Computers, vol. 1, no. 2, pp. 88–95, 2025.

[11] E. N. S. Wicaksana, R. Dijaya, I. A. Kautsar, and N. L. Azizah, "Perancangan Aplikasi Berbasis Web Untuk Membedakan Video Yang Dihasilkan Oleh (AI) Dengan Video Asli Menggunakan Metode Transformer," JATI (Jurnal Mahasiswa Teknik Informatika), vol. 10, no. 1, pp. 1615–1620, Feb. 2026.

[12] Z. Wang, Y. Chen, Y. Yao, M. Han, W. Xing, and M. Li, "IDCNet: Image Decomposition and Cross-View Distillation for Generalizable Deepfake Detection," IEEE Transactions on Information Forensics and Security, 2025.

[13] S. K. Panda, T. Diwan, O. G. Kakde, and J. V. Tembhurne, "Improvised Detection of Deepfakes From Visual Inputs Using Light Weight Deep Ensemble Model," Multimedia Tools and Applications, vol. 82, no. 13, pp. 20101–20118, 2023.

[14] S. Geetha Gowril, R. Devi, and K. Sethuraman, "MACHINE LEARNING," International Journal of Research and Analytical Reviews (IJRAR), vol. 6, no. 2, pp. 197-208, Jun. 2019.

[15] I. Bezzaoui, C. Stein, C. Weinhardt, and J. Fegert, "Explainable AI for online disinformation detection: Insights from a design science research project," Electronic Markets, vol. 35, no. 66, Jul. 2025, doi: 10.1007/s12525-025-00799-3.