Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{205365,
        author = {Ganga Ravi and Meera Nair and P L Nanditha Nair and Mahesh S},
        title = {PHISHIELD: Phishing Website Detection Using Machine Learning},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {1},
        pages = {6089-6094},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=205365},
        abstract = {The exponential growth of digital services has made phishing one of the most exploited attack vectors in cybersecurity, where threat actors deploy deceptive websites crafted to mimic legitimate platforms for the purpose of illicitly capturing user credentials, financial data, and sensitive personal information. Conventional defensive measures such as static blacklists and manually configured rule-based filters have proved inadequate against newly registered domains and continuously evolving phishing URLs that evade signature-based detection. This paper proposes PHISHIELD, an intelligent real-time phishing detection system built upon supervised machine learning, utilizing a curated feature set of 47 lexical and structural attributes extracted solely from URL strings — enabling swift classification without the computational overhead of full webpage rendering. Three classification models are trained, evaluated, and compared: Decision Tree, Support Vector Machine (SVM), and Gradient Boosting. Empirical evaluation conducted on a publicly available labelled phishing dataset demonstrates that the Gradient Boosting classifier achieves the highest detection accuracy of approximately 90%, surpassing competing models across precision, recall, and F1-score metrics. To strengthen resilience against zero-day threats beyond the scope of training data, PHISHIELD incorporates supplementary validation via the Google Safe Browsing API coupled with WHOIS-based domain-age interrogation. The system is operationalized as a Flask-based web application featuring a Bootstrap 5 responsive frontend and a secured REST API endpoint, facilitating seamless integration with browser extensions and enterprise-grade security infrastructures. The findings confirm that combining structured URL feature engineering, ensemble-based learning, and external threat intelligence yields a computationally lightweight, scalable, and practically deployable strategy for defending against modern phishing campaigns.},
        keywords = {Gradient Boosting, Machine Learning, Phishing Detection, Support Vector Machine, URL Feature Extraction, Web Security.},
        month = {June},
        }