Reliability and Robustness analysis of Machine Learning based Phishing URL Detectors
IEEE Transactions on Dependable and Secure Computing · 2024 · 1-18
DOI: 10.1109/tdsc.2022.3218043 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Dependable and Secure Computing, 2024. Verfasst von Bushra Sabir und 3 weiteren Autor:innen. Die Arbeit wurde bisher 13-mal zitiert (Crossref).
Abstract (Originalsprache)
—ML-based Phishing URL (MLPU) detectors serve as the first level of defence to protect users and organisations from being victims of phishing attacks. Lately, few studies havelaunchedsuccessfuladversarialattacksagainstspecificMLPUdetectorsraisingquestionsontheirpracticalreliabilityandusage.Nevertheless,therobustnessofthesesystems has not been extensively investigated. Therefore, the security vulnerabilities of these systems, in general, remain primarily unknown that calls for testing the robustness of these systems. In this article, we have proposed a methodology to investigate the reliability and robustness of 50 representative state-of-the-art MLPU models. Firstly, we have proposed a cost-effective Adversarial URL generator URLBUG that created an Adversarial URL dataset ( Adv data ) . Subsequently, we reproduced 50 MLPU (traditional ML and Deep learning) systems and recorded their baseline performance. Lastly, we tested the considered MLPU systems on Adv data and analyzed their robustness and reliability using box plots and heat maps. Our results showed that the generated adversarial URLs have valid syntax and can be registered at a median annual price of $11.99, and out of 13% of the already registered adversarial URLs, 63.94% were used for malicious purposes. Moreover, the considered MLPU models Matthew Correlation Coefficient (MCC) dropped from median 0.92 to 0.02 when tested against Adv data , indicating that the baseline MLPU models are unreliable in their current form. Further, our findings identified several security vulnerabilities of these systems and provided future directions for researchers to design dependable and secure MLPU systems.
Zitieren
APA
Sabir, B., Babar, M. A., Gaire, R., & Abuadbba, A. (2024). Reliability and Robustness analysis of Machine Learning based Phishing URL Detectors. IEEE Transactions on Dependable and Secure Computing, 1-18. https://doi.org/10.1109/tdsc.2022.3218043
Vancouver
Sabir B, Babar MA, Gaire R, Abuadbba A. Reliability and Robustness analysis of Machine Learning based Phishing URL Detectors. IEEE Transactions on Dependable and Secure Computing. 2024:1-18. doi:10.1109/tdsc.2022.3218043
BibTeX
@article{Sabir2024,
author = {Sabir, Bushra and Babar, M. Ali and Gaire, Raj and Abuadbba, Alsharif},
title = {Reliability and Robustness analysis of Machine Learning based Phishing URL Detectors},
journal = {IEEE Transactions on Dependable and Secure Computing},
year = {2024},
pages = {1--18},
doi = {10.1109/tdsc.2022.3218043}
}
EUVIMED