Advanced Evasion Attacks and Mitigations on Practical ML-Based Phishing Website Classifiers
This addresses security vulnerabilities in practical anti-phishing systems, which is critical for protecting users from cyber threats, though it is incremental on existing attack research.
The paper tackles evasion attacks on ML-based phishing website classifiers, showing that effective attacks are possible even with limited knowledge of the classifier while preserving webpage functionality and appearance, achieving 100% success rate on Google's filter and up to 81.25% on BitDefender's classifier. It also proposes a mitigation method called Pelican that can detect such attacks.
Machine learning (ML) based approaches have been the mainstream solution for anti-phishing detection. When they are deployed on the client-side, ML-based classifiers are vulnerable to evasion attacks. However, such potential threats have received relatively little attention because existing attacks destruct the functionalities or appearance of webpages and are conducted in the white-box scenario, making it less practical. Consequently, it becomes imperative to understand whether it is possible to launch evasion attacks with limited knowledge of the classifier, while preserving the functionalities and appearance. In this work, we show that even in the grey-, and black-box scenarios, evasion attacks are not only effective on practical ML-based classifiers, but can also be efficiently launched without destructing the functionalities and appearance. For this purpose, we propose three mutation-based attacks, differing in the knowledge of the target classifier, addressing a key technical challenge: automatically crafting an adversarial sample from a known phishing website in a way that can mislead classifiers. To launch attacks in the white- and grey-box scenarios, we also propose a sample-based collision attack to gain the knowledge of the target classifier. We demonstrate the effectiveness and efficiency of our evasion attacks on the state-of-the-art, Google's phishing page filter, achieved 100% attack success rate in less than one second per website. Moreover, the transferability attack on BitDefender's industrial phishing page classifier, TrafficLight, achieved up to 81.25% attack success rate. We further propose a similarity-based method to mitigate such evasion attacks, Pelican. We demonstrate that Pelican can effectively detect evasion attacks. Our findings contribute to design more robust phishing website classifiers in practice.