EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

The Effectiveness of early stopping on the efficiency of training CNN models for phishing URL identification

Muhammad Rifani Rifani, Septyan Eka Prastya, Muhammad Zulfadhilah, Munsyi

Install: Information System and Technology Journal · 08.05.2026 · 3 (1) · 1-9

DOI: 10.33859/install.v3i1.1024 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Install: Information System and Technology Journal, online erschienen am 08.05.2026. Verfasst von Muhammad Rifani Rifani und 3 weiteren Autor:innen. Die Arbeit nennt 15 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Phishing is a significant cybersecurity threat in which malicious URLs deceive users to steal sensitive data. Traditional detection methods, such as blacklists, often fail to keep pace with evolving phishing techniques. Deep learning, particularly Convolutional Neural Networks (CNNs), offers strong potential in phishing URL classification by capturing structural and semantic character-level patterns. However, CNN training demands high computational resources and risks overfitting. This study investigates the effectiveness of early stopping as a regularization technique to improve efficiency and generalization in character-based CNN models. Using a large-scale dataset of 130.080 URLs across four classes (benign, phishing, malware, defacement), the model employed character tokenization, embedding, convolution-pooling layers, and softmax classification. Early stopping monitored validation loss with patience values of 3, 5, and 10 epochs. Results show a 51% training time reduction and accuracy improvement from 96% to 97%, confirming early stopping as an efficient and robust detection approach.

Zitieren

APA

Rifani, M. R., Prastya, S. E., Zulfadhilah, M., & Munsyi (2026). The Effectiveness of early stopping on the efficiency of training CNN models for phishing URL identification. Install: Information System and Technology Journal, 3(1), 1-9. https://doi.org/10.33859/install.v3i1.1024

Vancouver

Rifani MR, Prastya SE, Zulfadhilah M, Munsyi. The Effectiveness of early stopping on the efficiency of training CNN models for phishing URL identification. Install: Information System and Technology Journal. 2026;3(1):1-9. doi:10.33859/install.v3i1.1024

BibTeX
@article{Rifani2026,
  author = {Rifani, Muhammad Rifani and Prastya, Septyan Eka and Zulfadhilah, Muhammad and Munsyi},
  title = {The Effectiveness of early stopping on the efficiency of training CNN models for phishing URL identification},
  journal = {Install: Information System and Technology Journal},
  year = {2026},
  volume = {3},
  number = {1},
  pages = {1--9},
  doi = {10.33859/install.v3i1.1024}
}
Kostenlos starten →