Phishing URL Detection and Interpretability With Machine Learning: A Cross‐Dataset Approach
SECURITY AND PRIVACY · 2026-01 · 9 (1)
DOI: 10.1002/spy2.70175 ↗Kurz gefasst
Fachartikel in der Zeitschrift SECURITY AND PRIVACY, online erschienen am 29.12.2025. Verfasst von Liyan Yi und 2 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 58 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Phishing attacks pose a significant security threat, particularly through deceptive emails designed to trick users into clicking on malicious links, with phishing URLs often serving as the primary indicator of such attacks. This paper presents a machine learning approach for detecting phishing email attacks by analyzing the URLs embedded within these emails, using Random Forest, eXtreme Gradient Boosting, and Light Gradient Boosting Machine models. Secondary datasets are used to evaluate model behavior and examine the applicability of model features across different samples. The models are assessed using metrics such as accuracy, precision, and recall to demonstrate their effectiveness in distinguishing between benign and malicious email URLs. The SHapley Additive exPlanations (SHAP) framework is employed to interpret the models' decision‐making processes and reinforce the relevance and reliability of key features. Our results show that across four test sets, the three models achieve an average classification error 4.03% and an average accuracy 94%, indicating strong generalization capability across diverse datasets using the same set of features.
Zitieren
APA
Yi, L., Omotosho, A., & Balogun, H. (2026). Phishing URL Detection and Interpretability With Machine Learning: A Cross‐Dataset Approach. SECURITY AND PRIVACY, 9(1). https://doi.org/10.1002/spy2.70175
Vancouver
Yi L, Omotosho A, Balogun H. Phishing URL Detection and Interpretability With Machine Learning: A Cross‐Dataset Approach. SECURITY AND PRIVACY. 2026;9(1). doi:10.1002/spy2.70175
BibTeX
@article{Yi2026,
author = {Yi, Liyan and Omotosho, Adebayo and Balogun, Hamed},
title = {Phishing URL Detection and Interpretability With Machine Learning: A Cross‐Dataset Approach},
journal = {SECURITY AND PRIVACY},
year = {2026},
volume = {9},
number = {1},
doi = {10.1002/spy2.70175}
}
EUVIMED