Artificial Intelligence and Machine Learning for Intelligent Phishing Website Detection - A Comprehensive Analysis and Future Research Directions

Phishing websites remain a significant cybersecurity challenge because attackers continuously modify their techniques to imitate legitimate websites and evade conventional detection mechanisms. This review examined the development of AI and ML-based phishing website detection, covering traditional machine learning, ensemble learning, deep learning, hybrid approaches, feature engineering, and Explainable Artificial Intelligence. The reviewed literature demonstrates a transition from static blacklist and rule-based detection toward intelligent approaches capable of analysing multiple website characteristics. Traditional machine learning remains valuable because of its relatively efficient classification capabilities, while ensemble and deep learning approaches provide additional capabilities for handling complex feature representations. Hybrid and multimodal approaches further provide opportunities to combine different sources of website evidence. However, the review also identifies important limitations. Existing research frequently depends on static datasets, inconsistent evaluation methodologies, and controlled experimental environments. Dataset diversity, model generalisation, computational efficiency, real-time detection, explainability, adaptive learning, and robustness against evolving and AI-assisted phishing techniques remain important challenges. The findings indicate that future phishing detection research should move beyond isolated improvements in classification performance and focus on systems that are adaptive, multimodal, explainable, computationally efficient, and suitable for real-world deployment. Standardised benchmarking and continuously updated datasets will also be important for evaluating the long-term effectiveness of detection models. Overall, AI and ML provide substantial potential for improving phishing website detection. However, achieving dependable real-world protection requires continued research into adaptability, transparency, robustness, scalability, and practical deployment. Addressing these challenges can contribute to the development of next-generation intelligent phishing website detection systems and strengthen cybersecurity in increasingly digital environments.  

[1]  J. Kumar, A. Santhanavijayan, B. Janet, B. Rajendran, and B. S. Bindhumadhava, “Phishing Website Classification and Detection Using Machine Learning,” in Proc. Int. Conf. Computational Collective Intelligence, 2020, doi: 10.1109/ICCCI48352.2020.9104161.  [2]   V. B. Wable and P. Thakur, “CheckPhish: Leveraging a Machine Learning Approach for Detecting Phishing Websites,” in Proc. IEEE Int. Conf. Computing, Communication and Networking Technologies, 2023, doi: 10.1109/ICCCNT56998.2023.10308156. [3]    A. Kara, A. Ok, and A. Ozaday, “Characteristics of Understanding URLs and Domain Names Features: The Basis for Phishing Detection,” IEEE Access, 2022, doi: 10.1109/ACCESS.2022.3223111.  [4]    “Detection of Phishing Websites Using an Efficient Feature-Based Machine Learning Framework,” International Journal of Engineering and Advanced Technology, 2020, doi: 10.35940/ijeat.C5909.029320.  [5]   H. Kansagara et al., “A Hybrid Approach for Phishing Website Detection Using Machine Learning,” 2021.  [6]    K. Omari, “Comparative Study of Machine Learning Algorithms for Phishing Website Detection,” International Journal of Advanced Computer Science and Applications, 2023, doi: 10.14569/IJACSA.2023.0140945.  [7]   “Phishing Website Detection Using Deep Learning,” in Proc. ACM International Conference, 2022, doi: 10.1145/3542954.3542967.  [8]  “Analysis of Phishing Website Detection Using CNN and Bidirectional LSTM,” in Proc. International Conference on Electronics, Communication and Aerospace Technology, 2020, doi: 10.1109/ICECA49313.2020.9297395.  [9]   D.-J. Liu et al., “AntiPhishNN: Hybrid Neural Networks Using URL and Domain Information for Phishing Detection,” 2023.  [10]    “A Deep Learning-Based Innovative Technique for Phishing Detection in URLs,” Sensors, vol. 23, no. 9, 2023, doi: 10.3390/s23094403.