EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A feature-engineered dataset of benign and phishing URLs for machine learning and large language models evaluation

Dam Minh Linh, Tran Cong Hung

Data in Brief · 2025-12 · 63 · 112162

DOI: 10.1016/j.dib.2025.112162 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Data in Brief, 2025-12. Verfasst von Dam Minh Linh und 1 weiteren Autor:innen. Beteiligte Einrichtung: Information Security Technology Laboratory, and Faculty of Information Technology, Posts and Telecommunications Institute of Technology (PTIT), Ho Ch…. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 12 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Posts and Telecommunications Institute of Technology.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Phishing websites remain a major cybersecurity threat, yet the availability of balanced and feature-rich datasets for evaluating detection models is still limited. While machine learning (ML) and large language models (LLMs) have shown strong potential in URL-based classification, most public datasets provide raw URLs without feature engineering, making reproducibility and fair comparison across models difficult. To address this gap, we present a curated dataset of 111,660 URLs, consisting of 100,000 benign samples (label 0) and 11,660 phishing samples (label 1). Each URL entry is enriched with 22 numerical lexical and structural features (e.g., URL length, domain length, digit ratio, entropy, HTTPS usage). Additionally, three string reference columns (URL, domain, TLD) are preserved for interpretability, and one label column (0 = benign, 1 = phishing), totaling 26 columns. To demonstrate its utility, we evaluate two baseline approaches: a Random Forest (RF) classifier using handcrafted features, and a MiniLM embedding model with Logistic Regression (LR). Both achieved accuracy above 96 % and ROC AUC scores exceeding 0.99 across training, validation, and test splits. This dataset represents an important step toward building reproducible and comparable benchmarks for phishing detection, bridging traditional ML and LLM-based approaches, and supporting future research on adversarial robustness and scalable security models.

Schlagwörter

Stichwörter der Autor:innen

Artificial intelligence (AI)CybersecurityData scienceFeature-engineered datasetLarge language models (LLMs)Machine learning (ML)Natural language processing (NLP)URL classification

Autor:innen und Institutionen

  • Information Security Technology Laboratory, and Faculty of Information Technology, Posts and Telecommunications Institute of Technology (PTIT), Ho Chi Minh, Vietnam.
  • Faculty of Engineering and Computer Science, Saigon International University (SIU), Ho Chi Minh, Vietnam.

Förderung

  • Posts and Telecommunications Institute of Technology

Zitieren

APA

Linh, D. M., & Hung, T. C. (2025). A feature-engineered dataset of benign and phishing URLs for machine learning and large language models evaluation. Data in Brief, 63, 112162. https://doi.org/10.1016/j.dib.2025.112162

Vancouver

Linh DM, Hung TC. A feature-engineered dataset of benign and phishing URLs for machine learning and large language models evaluation. Data in Brief. 2025;63:112162. doi:10.1016/j.dib.2025.112162

BibTeX
@article{Linh2025,
  author = {Linh, Dam Minh and Hung, Tran Cong},
  title = {A feature-engineered dataset of benign and phishing URLs for machine learning and large language models evaluation},
  journal = {Data in Brief},
  year = {2025},
  volume = {63},
  pages = {112162},
  doi = {10.1016/j.dib.2025.112162}
}
Kostenlos starten →