An Ensemble Machine Learning Approach with Custom-Refined Features for Detecting Phishing URLs and Preventing Cyber Attack
The Asian Bulletin of Big Data Management · 31.03.2026 · 6 (1) · 598-621
DOI: 10.62019/2p4jyc15 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Asian Bulletin of Big Data Management, online erschienen am 31.03.2026. Verfasst von Renum Raza und 4 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
Phishing attacks remain a major cybersecurity hazard by masquerading as trustworthy websites to trick users into disclosing sensitive data. This research introduces a web-based Phishing URL Detection System that applies machine learning to reliably classify URLs as either benign or malicious. Built with Python, Flask, and core machine learning libraries, the platform is trained on a rich dataset that includes lexical, content, and structural features drawn from the URLs. The model uses a compact subset of 20 refined lexical and content-based features derived from the 56-feature Kaggle dataset through correlation filtering and rule-based parsing. This study proposes a soft-voting ensemble model combining Random Forest, Support Vector Machine, and Decision Tree classifiers for phishing URL detection. The ensemble leverages complementary decision boundaries of individual models to improve generalization and predictive performance. Experimental results demonstrate that the proposed ensemble outperforms individual classifiers across all evaluation metrics and is deployed in a real-time web-based detection system. The application incorporates a secure user authentication module and a straightforward interface where users can submit URLs and receive immediate hazard evaluations. Moreover, the system identifies the specific features driving each classification, educating users on recognizable indicators of phishing schemes. The overall goal is to enhance online safety by providing a trustworthy, transparent, and approachable tool for timely phishing detection.
Zitieren
APA
Raza, R., Sattar, H., Shafi, U. F., Zahrah, & Gillani, M. (2026). An Ensemble Machine Learning Approach with Custom-Refined Features for Detecting Phishing URLs and Preventing Cyber Attack. The Asian Bulletin of Big Data Management, 6(1), 598-621. https://doi.org/10.62019/2p4jyc15
Vancouver
Raza R, Sattar H, Shafi UF, Zahrah, Gillani M. An Ensemble Machine Learning Approach with Custom-Refined Features for Detecting Phishing URLs and Preventing Cyber Attack. The Asian Bulletin of Big Data Management. 2026;6(1):598-621. doi:10.62019/2p4jyc15
BibTeX
@article{Raza2026,
author = {Raza, Renum and Sattar, Hina and Shafi, Umar Farooq and Zahrah and Gillani, Mariam},
title = {An Ensemble Machine Learning Approach with Custom-Refined Features for Detecting Phishing URLs and Preventing Cyber Attack},
journal = {The Asian Bulletin of Big Data Management},
year = {2026},
volume = {6},
number = {1},
pages = {598--621},
doi = {10.62019/2p4jyc15}
}
EUVIMED