Multi-SpacePhish: Extending the Evasion-space of Adversarial Attacks against Phishing Website Detectors using Machine Learning
This work addresses the lack of secure ML systems by clarifying the actual feasibility and threat of adversarial attacks for phishing detection, which is an incremental but important step for cybersecurity applications.
The paper tackles the problem of evaluating the real-world threat of adversarial attacks on phishing website detectors by formalizing the evasion-space and proposing realistic threat models. It shows that realistic evasion attempts cause a statistically significant degradation of 3-10% in detection rates, with some detectors remaining immune, and multi-space perturbations can drop detection rates from 0.95 to 0.
Existing literature on adversarial Machine Learning (ML) focuses either on showing attacks that break every ML model, or defenses that withstand most attacks. Unfortunately, little consideration is given to the actual feasibility of the attack or the defense. Moreover, adversarial samples are often crafted in the "feature-space", making the corresponding evaluations of questionable value. Simply put, the current situation does not allow to estimate the actual threat posed by adversarial attacks, leading to a lack of secure ML systems. We aim to clarify such confusion in this paper. By considering the application of ML for Phishing Website Detection (PWD), we formalize the "evasion-space" in which an adversarial perturbation can be introduced to fool a ML-PWD -- demonstrating that even perturbations in the "feature-space" are useful. Then, we propose a realistic threat model describing evasion attacks against ML-PWD that are cheap to stage, and hence intrinsically more attractive for real phishers. After that, we perform the first statistically validated assessment of state-of-the-art ML-PWD against 12 evasion attacks. Our evaluation shows (i) the true efficacy of evasion attempts that are more likely to occur; and (ii) the impact of perturbations crafted in different evasion-spaces. Our realistic evasion attempts induce a statistically significant degradation (3-10% at p<0.05), and their cheap cost makes them a subtle threat. Notably, however, some ML-PWD are immune to our most realistic attacks (p=0.22). Finally, as an additional contribution of this journal publication, we are the first to consider the intriguing case wherein an attacker introduces perturbations in multiple evasion-spaces at the same time. These new results show that simultaneously applying perturbations in the problem- and feature-space can cause a drop in the detection rate from 0.95 to 0.