An Explainable Ensemble Learning Framework for Real-Time Phishing Email Detection

This study proposes an Explainable Ensemble Learning Framework for Real-Time Phishing Email Detection using NLP, Machine Learning, Ensemble Classification, SHAP Explainability Analysis and Flask based real-time detection interface. The suggested approach was able to effectively identify phishing emails via intelligent textual feature analysis and explainable machine learning techniques. Experimental evaluation showed good classification accuracy, less false positives, improved robustness and transparency. The incorporation of SHAP explainability increased the trustworthiness and interpretability of phishing prediction decisions. The suggested architecture offers an intelligent, scalable, explainable and practical solution for modern phishing email detection systems and contributes towards enhancing cyber security against evolving phishing attacks.

[1] R. Mahajan and I. Siddavatam, “Phishing Detection using Machine Learning Techniques,” International Journal of Computer Applications, vol. 181, no. 23, pp. 45–50, 2018. [2] S. Lundberg and S. Lee, “A Unified Approach to Interpreting Model Predictions,” Advances in Neural Information Processing Systems, pp. 4765–4774, 2017. [3] T. Mikolov et al., “Efficient Estimation of Word Representations in Vector Space,” arXiv preprint arXiv:1301.3781, 2013. [4] S. Hochreiter and J. Schmidhuber, “Long Short-Term Memory,” Neural Computation,vol. 9, no. 8, pp. 1735–1780, 1997. [5] T. Chen and C. Guestrin, “XGBoost: A Scalable Tree Boosting System,” Proc. ACM SIGKDD,pp. 785–794, 2016. [6] F. Pedregosa et al.,“Scikit-learn: Machine Learning in Python,” Journal of Machine Learning Research, vol. 12, pp. 2825–2830, 2011. [7] Apache SpamAssassin, “Public Corpus Dataset,” Available: https://spamassassin.apache.org/ [8] Enron Email Dataset, Available: https://www.cs.cmu.edu/~enron/ [9] Nazario Phishing Corpus, Available: https://monkey.org/~jose/phishing/ [10] D. Dua and C. Graff, “UCI Machine Learning Repository,” University of California, Irvine, 2019.