Yazar "Farahmandnia, Feraidoon" seçeneğine göre listele

Listeleniyor 1 - 1 / 1
  • Küçük Resim
    Öğe
    Enhanced DDoS attack detection through hybrid machine learning techniques
    (İstanbul Ticaret Üniversitesi, 2025) Farahmandnia, Feraidoon; Özekes, Serhat
    This study investigates the development of robust detection mechanisms for Distributed Denial of Service (DDoS) attacks using machine learning techniques. The primary objective of the research is to enhance DDoS detection accuracy by exploring two distinct approaches: a meta-classifier stacking model and a transfer learning model. The CICDDoS2019 and CICIDS2017 datasets are utilized to train and evaluate these models. In the first approach, the K-Nearest Neighbors, Support Vector Machine, and Random Forest algorithms are combined using a logistic regression metaclassifier. This ensemble method leverages the strengths of each individual algorithm, resulting in improved performance metrics such as accuracy, precision, recall, and F1-score. The stacking model achieved an accuracy of 99.94%. The second approach employs transfer learning, where a pre-trained Artificial Neural Network model on the CICIDS2017 is fine-tuned using the CICDDoS2019 dataset. This method demonstrates the advantages of knowledge transfer, achieving high detection performance with an accuracy of 99.81% and significantly reduced training time. The findings indicate that both approaches significantly improve DDoS detection. The metaclassifier approach achieves higher performance metrics but is more computationally intensive. The transfer learning approach offers a practical balance between performance and efficiency, making it suitable for scenarios requiring rapid model deployment. In conclusion, the research highlights the potential of advanced machine learning techniques in developing effective DDoS detection systems.