Answers are grounded in this paper and cite their sources.
Review highlights advancements in phishing url detection using machine learning and deep learning, indicating key challenges and future directions.
Phishing has emerged as a prevalent security threat, relying on deceptive practices to trick individuals and organizations into divulging sensitive information. As phishing techniques grow more advanced, traditional defensive strategies often prove insufficient for detecting new or cleverly disguised attacks. This paper presents a thorough review of phishing URL detection methods, charting the evolution from blacklist-based and heuristic methods to contemporary approaches grounded in machine learning (ML), deep learning (DL), and explainable artificial intelligence (XAI). We also analyze the challenges these systems face in real-world applications, including adversarial attacks, scalability concerns, and the necessity of privacy-preserving mechanisms such as federated learning. Finally, we propose future directions aimed at bolstering detection performance and interpretability to empower comprehensive cybersecurity solutions against phishing.
No takes yet. Share an insight, caveat, or question.
Lambture et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: