Adversarial threat evaluation of machine learning-based phishing detectors: a cybersecurity-focused black-box perspective

Key Points

• The study aims to evaluate the effectiveness of machine learning-based phishing detectors against adversarial threats.
• Identified types of adversarial attacks targeting phishing detectors.
• Conducted evaluations of various machine learning algorithms under simulated attack scenarios.
• Analyzed detection accuracy and response to specific adversarial manipulations.
• Found significant vulnerabilities in existing phishing detection systems.
• Demonstrated that certain attack methods considerably reduced detection rates.
• Identified potential strategies to enhance resilience against adversarial threats.