Browsing by Author "Daǧ, H."

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • Thumbnail Image
    Conference Object
    Citation - Scopus: 0
    Resource-Efficient Ensemble Learning for Edge Iiot Network Security Against Osint-Based Attacks
    (Institute of Electrical and Electronics Engineers Inc., 2024) Ecevit, M.I.; Çukur, Z.; Izgün, M.A.; Ui Ain, N.; Daǧ, H.
    The rise of Edge IIoT networks has transformed industries by enabling real-time data processing, but these networks face significant c ybersecurity risks, particularly from OSINT-based attacks. This paper presents a resource-efficient ensemble learning framework designed to detect such attacks in Edge IIoT environments. The framework integrates machine learning models, including RandomForest, K-Nearest Neighbors, and Logistic Regression, optimized with Principal Component Analysis (PCA) to reduce data dimensionality and computational overhead. GridSearchCV and StratifiedKFold cross-validation were employed to fine-tune the models, resulting in high detection accuracy. This approach ensures robust and efficient security for resource-constrained Edge IIoT networks. © 2024 IEEE.
  • Thumbnail Image
    Conference Object
    Citation - Scopus: 0
    Transfer Learning for Phishing Detection: Screenshot-Based Website Classification
    (Institute of Electrical and Electronics Engineers Inc., 2024) Çolhak, F.; Ecevit, M.I.; Daǧ, H.
    Phishing remains a significant threat in the evolving cybersecurity landscape as phishing websites become increasingly similar to legitimate websites, complicating detection using traditional methods. This study explores AI-based solutions for screenshot-based phishing detection, utilizing the MTLP dataset and applying transfer learning with pretrained models (DenseNet, ResNet, EfficientNet, Inception, MobileNet, VGG) using the timm library. The study also discusses challenges related to phishing datasets and compares publicly available datasets, highlighting MTLP Dataset's strengths. DenseNetBlur121D was identified as the top-performing model, achieving an accuracy of 95.28%, a recall of 95.38%, a precision of 93.42%, and an F1 score of 94.39% when applied to the entire MTLP dataset. Both the model code and dataset are publicly available, providing a valuable resource for further research and development in this domain. © 2024 IEEE.