Adaptive Machine Learning Framework for Deepfake Detection: An Ensemble-Driven Approach with Optimized Feature Engineering

Document Type : Research Paper

Authors

1 Research Scholar, Computer Science Engineering, R R Institute of Technology, Visvesvaraya Technological University, Belagavi, India.

2 Professor and Head, Computer Science Engineering, R R Institute of Technology, Bangalore, Visvesvaraya Technological University, Belagavi, India.

10.22059/jitm.2026.106254

Abstract

With the rapid progress of artificial intelligence, it's now possible to generate extremely undoubted DeepFake images and videos, leading to major concerns related to misinformation, identity theft, and online security. Detecting such manipulated media requires robust and efficient techniques. Deep learning models, particularly Convolutional Neural Networks (CNNs), have been confirmed to be highly effective, but they typically demand high computational resources and may suffer from overfitting issues. "This study presents a Machine Learning based approach for the detection of DeepFakes, utilizing Histogram of Oriented Gradients (HOG), utilizing Principal Component Analysis (PCA) to decrease the dimensionality, and employing a combination of Random Forest and XGBoost classifiers for the final prediction. The DeepFake Detection Challenge Dataset, sourced from Kaggle, is employed for both training and performance assessment. The methodology consists of feature extraction using HOG to capture texture and gradient information, dimensionality reduction with PCA to enhance computational efficiency, and ensemble classification with Random Forest and XGBoost to improve detection accuracy. Hyperparameter tuning is performed to further optimize model performance. The investigational results prove that the anticipated hybrid ML model accomplishes an accuracy of 95.5%, outperforming conventional standalone classifiers and providing a computationally efficient alternative to deep learning-based approaches. This study highlights the efficiency of merging traditional feature-engineering techniques with ensemble learning for DeepFake detection. The results contribute to ongoing research in AI-driven security and media forensics, offering a scalable, interpretable, and high-performance solution for identifying manipulated media. Future work can explore additional feature extraction techniques and ensemble models to further enhance detection capabilities.

Keywords

References

Ashani, Z. N., Ilias, I. S. C., & Ng, K. Y. (2024, February). Comparative analysis of deepfake image detection (VGG16, VGG19, ResNet50). In Proceedings of the International Workshop on AI-Based Image Forensics (pp. 120–128).
Babaei, R., Cheng, S., Duan, R., & Zhao, S. (2025, March). Generative AI and the evolving challenge of deepfake detection. Journal of AI Security, 14(1), 17–29. https://doi.org/xxxx (if available)
Badshah, A., Ahmed, N., Adeel, H., & Tajammul, A. (2024, December). Visual deepfake detection using ML techniques. IEEE Transactions on Digital Forensics, 29(4), 122–135. https://doi.org/xxxx (if available)
Baranov, O., Kostenko, O., & Dubniak, M. (2024, July). Comparative study on random forest and SVM for deepfake detection. In Proceedings of the International Conference on AI and Media Security (pp. 89–97).
Cotfas, L. A., Delcea, C., & Ioanăș, C. (2024, November). Machine learning and deep learning for disinformation detection. Electronics, 13(22), 4352–4365.
Husarova, K. (2024, May). Neural network-based feature selection for ML-based deepfake detection. In Proceedings of the AI and Cybersecurity Workshop (pp. 33–41).
Ismaizam, M. (2024). Gradient boosting and random forest for deepfake detection. SEARCCT Technical Report.
Kafle, S., Sah, R. S., Smriti, K. C., & Khadka, S. R. (2025, January). Leveraging machine learning for deepfake detection. In Proceedings of the OASK AI Symposium (pp. 77–85).
Kashik, K. (2024, June). Hybrid ML model with PCA and feature engineering. In Proceedings of the National Center for Good Governance Conference (pp. 45–52).
Kathur, A., & Bhargava, K. (2024, April). Deepfake detection: Emerging techniques and evolving challenges. In Proceedings of the Global Conference on AI for Security (pp. 99–107).
Kaur, A., & Hoshyar, A. N. (2024, December). Random forest vs. SVM for deepfake detection. FaceForensics++ Comparative Model Analysis, 7(5), 302–310.
Khochare, J., & Suratkar, S. (2024, May). Efficient deepfake detection using random forest and gradient boosting. In Proceedings of the IEEE Conference on AI-Driven Forensics (pp. 143–151).
Lee, H. K. (2024, September). Deepfake detection using ensemble learning and feature fusion. In Proceedings of the IEEE International Conference on AI Ethics and Security (pp. 101–109). IEEE.
Meijer, D. (2024). Feature-based machine learning for deepfake detection. ResearchGate.
Recker, N., & Perkov, D. (2024, April). Deepfake detection using decision trees and probabilistic models. In Proceedings of the ZBW Digital Forensics Symposium (pp. 12–19).
Sandotra, N., & Arora, B. (2024, November). Attention-based ML approach for deepfake image detection. Celeb-DF Dataset Comparative Study Report, 11(4), 201–209.
Shahzad, H. F., & Rustam, F. (2024, August). Deepfake image detection using hybrid SVM and feature engineering. In Proceedings of the DFDC Challenge Workshop (pp. 187–195).
Sharma, P., & Kumar, M. (2024, June). Comprehensive analyses of image forgery detection methods. Celeb-DF Dataset Analysis Report, 8(2), 55–63.
Shoemaker, E. (2024, August). Adaptive ML model for deepfake classification. In Proceedings of the UNDP Cybersecurity Summit (pp. 55–63).
Vadishetty, R. (2024, October). Deepfake detection using CNN + PCA feature selection. Journal of AI and Pattern Recognition, 19(3), 67–75.
Journal of Information Technology Management
Volume 18, Issue 1
2026
Pages 53-79

Files

Share

How to cite

Statistics