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{200935,
        author = {Mr. T. Kamalesh and Arunkumar RS and Dinakaran K and Mohanraj R},
        title = {Hybrid Machine Learning and Deep Learning Framework for Accurate Phishing Website Detection with AI Model Integration},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {no},
        pages = {44-54},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=200935},
        abstract = {Phishing attacks represent one of the most pervasive and financially devastating cyber threats in the modern digital landscape, compromising millions of users annually by employing deceptive websites engineered to mimic legitimate online services. Conventional detection mechanisms, including static blacklists and heuristic rule-based filters, exhibit critical limitations in identifying zero-day phishing campaigns and dynamically evolving attack vectors. This paper presents a novel Hybrid Machine Learning and Deep Learning Framework integrated with a Large Language Model (LLM)-powered AI Explanation Module, designed to deliver accurate, real-time phishing website classification with contextual user awareness. The proposed system processes user-submitted URLs through a multi-stage analytical pipeline encompassing URL preprocessing, WHOIS domain lookup, QR code decoding, and comprehensive feature engineering across more than one hundred lexical, domain-centric, and content-based attributes. Classification is performed using an ensemble of Random Forest, Support Vector Machine (SVM), and Decision Tree algorithms, augmented by a neural network layer for enhanced generalization. Experimental evaluation conducted on a labeled dataset of approximately 90,000 URLs demonstrates that the Random Forest classifier achieves peak performance with 96% accuracy, 94% precision, 95% recall, and an F1-score of 95%, alongside an AUC of 0.99. The system further incorporates an AI-driven explanation module that elucidates detection rationale, identifies specific risk factors, and provides actionable cybersecurity guidance to end users. The web-based deployment ensures cross-platform accessibility on both mobile and desktop environments, positioning the framework as a practical, scalable solution for real-time phishing defense.},
        keywords = {Phishing detection, machine learning, deep learning, Random Forest, feature engineering, URL analysis, cybersecurity, AI explanation, neural network, WHOIS lookup.},
        month = {May},
        }