Phishing remains to be a key issue in terms of online security as it uses quite misleading URLs and continuously
changing attack mechanisms in order to avoid the classical methods of detection. In this research, a minimalistic AI-boosted
phishing identification system that utilizes machine learning (ML) algorithms on classifying phishing URLs in real-time is
proposed. The suggested system uses a combination of two high-quality repositories, namely PhishTank and the University of
New Brunswick (UNB) and extracts eleven-one wildly various features, such as lexical, host, and content features. The
effectiveness of the eight machine learning models, which we benchmark, are Logistic Regression (LR), Decision Tree(DT),
Random Forest (RF), Naive Bayes (NB), Support Vector Machine (SVM), XGBoost, LightGBM, and a Multilayer Perceptron
(MLP). Evaluation of these models is carried out by a standard measure of performance like accuracy, precision, recall, F1
score and efficiency. According to the experiment, ensemble and gradient boosting models can be regarded as more
successful than others, and LightGBM can be noted as a good choice when it comes to the use in real-time because it
maximizes the speed and accuracy in both tests. In general, the research is low-cost, scalable and non-technical in nature to
know the evolution in the new improved web security regarding the advanced machine learning practices.
[1] N. Altwaijry, I. Al-Turaiki, R. Alotaibi, and F. Alakeel, "Advancing Phishing Email Detection: A Comparative Study of
Deep Learning Models," Sensors, vol. 24, no. 7, p. 2077, Mar. 2024.
[2] S. Aslam, H. Aslam, A. Manzoor, C. Hui, and A. Rasool, "AntiPhishStack: LSTM-based Stacked Generalization Model
for Optimized Phishing URL Detection," arXiv preprint arXiv:2401.08947, Jan. 2024.
[3] W. Guo, Q. Wang, H. Yue, H. Sun, and R. Q. Hu, "Efficient Phishing URL Detection Using Graph-based Machine
Learning and Loopy Belief Propagation," arXiv preprint arXiv:2501.06912, Jan. 2025.
[4] M. A. Daniel, S.-C. Chong, L.-Y. Chong, and K.-K. Wee, "Optimising Phishing Detection: A Comparative Analysis of
Machine Learning Methods with Feature Selection," Journal of Informatics and Web Engineering, vol. 4, no. 1, pp. 200–212,
Feb. 2025.
[5] P. An, R. Shafi, T. Mughogho, and O. A. Onyango, "Multilingual Email Phishing Attacks Detection using OSINT and
Machine Learning," arXiv preprint arXiv:2501.08723, Jan. 2025.
[6] R. Meléndez, M. Ptaszynski, and F. Masui, "Comparative Investigation of Traditional Machine-Learning Models and
Transformer Models for Phishing Email Detection," Electronics, vol. 13, no. 24, p. 4877, Dec. 2024.
[7] R. A. Khurma, K. E. Sabri, P. A. Castillo, and I. Aljarah, "Hybrid Phishing Detection Based on Automated Feature
Selection Using the Chaotic Dragonfly Algorithm," Electronics, vol. 12, no. 13, p. 2823, Jul. 2023.
[8] P. Maneriker et al., "URLTran: Improving Phishing URL Detection Using Transformers," arXiv preprint
arXiv:2106.05256, June 2021.
[9] A. Makkar, N. Kumar, L. Sama, S. Mishra, and Y. Samdani, "An Intelligent Phishing Detection Scheme Using Machine
Learning," in Proc. Sixth Int. Conf. Mathematics and Computing, Springer, Singapore, 2021, pp. 131–142.
[10] A. Al Darmaki, M. Bait-Suwailam, A. Khan, and M. R. Mughal, "Phishing Detection Simulations: A Comparative
Analysis of Machine Learning Models," Engineering Research Express, 2024.
