An Explainable and Feature-Efficient Machine-Learning Framework for Real-Time Phishing URL Detection

This paper proposes an explainable and feature-efficient framework for real-time phishing URL detection. The design combines lexical and structural URL features, four classical machine-learning classifiers, training-fold-only feature selection, SHAP/LIME or intrinsic explanations, and explicit measurement of discrimination, false positives, resource use, and latency. The protocol is designed to reduce data leakage and expose whether compact models generalize beyond a random split. The numerical values retained in Section VIII should be treated as provisional until the dataset, implementation, and evaluation pipeline are executed and independently checked. Final deployment claims should therefore be based on held-out testing, robustness analysis, latency measurements, and reproducibility evidence

[1] S. S. Shafin, “An Explainable Feature Selection Framework for Web Phishing Detection with Machine Learning,” 2024. [2] N. Pundlik, T. Saxena, and J. Mungi, “A Scalable Explainable AI Framework for Phishing Attack Detection with Interpretable Feature Insights with Machine Learning Based Ensemble Models,” 2025. [3] D. R. Patil et al., “Enhanced Phishing URLs Detection using Feature Selection and Machine Learning Approaches,” 2024. [4] R. A. D. Talasari and E. A. Kusnanti, “Phishing URL Detection Using URL Characteristics: A Multi-Algorithm Machine Learning and Explainable AI (XAI) Approach,” 2026. [5] H. K. M. Lai et al., “Explainable Machine Learning Framework for Phishing URL Detection,” 2026. [6] R. Alam et al., “E2Phish: Explainable Ensemble Machine Learning Model for Enhanced Phishing URL Detection,” 2024. [7] M. Alajlouni et al., “Optimizing Phishing URL Detection: A Comparative Analysis of Feature Counts and Training Times,” 2024. [8] L. M. Rani et al., “Feature Selection to Enhance Phishing Website Detection Based On URL Using Machine Learning Techniques,” 2023. [9] M. AbuTaha et al., “URL Phishing Detection using Machine Learning Techniques based on URLs Lexical Analysis,” 2021. [10] B. Banik et al., “Lexical Feature Based Feature Selection and Phishing URL Classification Using Machine Learning Techniques,” 2020. [11] S. Ali et al., “Entropy-Based Feature Selection Classification Approach for Detecting Phishing Websites,” 2019.