Enhancing Phishing Detection Using BERT and Graph Neural Network Approach
International Journal of Innovative Science and Research Technology · 05.06.2026 · 3244
DOI: 10.38124/ijisrt/26may1421 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Innovative Science and Research Technology, online erschienen am 05.06.2026. Verfasst von Zainab Jibril Amedu und 2 weiteren Autor:innen. Die Arbeit nennt 74 Literaturangaben.
Abstract (Originalsprache)
This paper presents a novel hybrid deep learning architecture for phishing detection that integrates BERT and Graph Neural Networks through cross-modal attention fusion. The proposed model addresses the multimodal nature of phishing attacks by simultaneously processing textual features via DistilBERT and structural relationships via a Heterogeneous Graph Transformer. Our methodology employs a security-aware loss function emphasizing false positive reduction and implements 5-fold cross-validation for robust evaluation.
Zitieren
APA
Amedu, Z. J., Kirubakaran, P., & Kolapo, R. (2026). Enhancing Phishing Detection Using BERT and Graph Neural Network Approach. International Journal of Innovative Science and Research Technology, 3244. https://doi.org/10.38124/ijisrt/26may1421
Vancouver
Amedu ZJ, Kirubakaran P, Kolapo R. Enhancing Phishing Detection Using BERT and Graph Neural Network Approach. International Journal of Innovative Science and Research Technology. 2026:3244. doi:10.38124/ijisrt/26may1421
BibTeX
@article{Amedu2026,
author = {Amedu, Zainab Jibril and Kirubakaran, Prema and Kolapo, Ridwan},
title = {Enhancing Phishing Detection Using BERT and Graph Neural Network Approach},
journal = {International Journal of Innovative Science and Research Technology},
year = {2026},
pages = {3244},
doi = {10.38124/ijisrt/26may1421}
}
EUVIMED