Comparative Analysis of Transformer-Based and Classical Machine Learning Models for Phishing Email Detection: A Multi-Source Dataset Evaluation with Explainability
Enfoque UTE · 01.10.2026 · 17 (4) · 1-11
DOI: 10.29019/enfoqueute.1284 ↗Kurz gefasst
Fachartikel in der Zeitschrift Enfoque UTE, online erschienen am 01.10.2026. Verfasst von Andre Sebastian Samaniego Buñay und 2 weiteren Autor:innen. Die Arbeit nennt 16 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Phishing remains one of the most persistent cyber threats, particularly in email environments where deceptive messages can be distributed at scale. This paper compares five classifiers: Multinomial Naive Bayes, Random Forest, Bidirectional Long Short-Term Memory (BiLSTM), DistilBERT, and BERT-base. A multi-source corpus of 82,689 cleaned and deduplicated emails was built from nine public datasets. Under a unified protocol, BERT-base achieved the highest F1-score (0.9824), while DistilBERT obtained an almost identical F1-score (0.9822) with lower measured inference latency (1.130 versus 2.218 ms/email), representing the strongest accuracy–latency trade-off in the evaluated environment. LIME explanations exposed plausible phishing indicators, such as urgency and account-verification language, but also mixed local contributions that require cautious interpretation. In the source-held-out experiment, the positive-class prevalence changed from 41.1% in training to 26.6% in testing, and DistilBERT produced 505 false negatives but only two false positives. Consequently, recall decreased from 0.9750 to 0.5346, showing that high mixed-source test performance does not guarantee robustness when complete data sources are unseen.
Zitieren
APA
Buñay, A. S. S., Albarracin, A. M. O., & Pomagualli, J. M. C. (2026). Comparative Analysis of Transformer-Based and Classical Machine Learning Models for Phishing Email Detection: A Multi-Source Dataset Evaluation with Explainability. Enfoque UTE, 17(4), 1-11. https://doi.org/10.29019/enfoqueute.1284
Vancouver
Buñay ASS, Albarracin AMO, Pomagualli JMC. Comparative Analysis of Transformer-Based and Classical Machine Learning Models for Phishing Email Detection: A Multi-Source Dataset Evaluation with Explainability. Enfoque UTE. 2026;17(4):1-11. doi:10.29019/enfoqueute.1284
BibTeX
@article{Buay2026,
author = {Buñay, Andre Sebastian Samaniego and Albarracin, Ariel Misael Orellana and Pomagualli, Joel Marcelo Chuquimarca},
title = {Comparative Analysis of Transformer-Based and Classical Machine Learning Models for Phishing Email Detection: A Multi-Source Dataset Evaluation with Explainability},
journal = {Enfoque UTE},
year = {2026},
volume = {17},
number = {4},
pages = {1--11},
doi = {10.29019/enfoqueute.1284}
}
EUVIMED