Detecting Phishing Websites Using Transformer-Based Language Models and URL Features

Authors

  • Al-Ahmadi. S., AbuSharkh. M Author

Keywords:

Phishing detection, transformer, URL analysis, language models, feature fusion, cybersecurity, classification

Abstract

Phishing websites use deceptive links to direct users to fraudulent pages and obtain sensitive information. This paper proposes a hybrid phishing detection framework that combines transformer-based language models with engineered URL features. A transformer encoder processes URL tokens to learn contextual and sequential patterns, while a feature extraction module calculates lexical and structural properties such as URL length, symbol frequency, domain composition, and suspicious token patterns. The two representations are fused and passed to a classification layer that predicts whether a URL is legitimate or phishing-related. The proposed workflow includes data preparation, model training, evaluation, and a decision stage for classifying incoming URLs. The framework is designed to support early detection without requiring full webpage rendering. Its effectiveness must be established through experiments using separate training, validation, and test data, comparison with baseline models, and evaluation using precision, recall, F1-score, and false-positive rate.

References

Maneriker, P., Stokes, J. W., Garcia Lazo, E., Carutasu, D., Tajaddodianfar, F., & Gururajan, A. (2021). URLTran: Improving phishing URL detection using transformers. MILCOM 2021—2021 IEEE Military Communications Conference, 197–204. https://doi.org/10.1109/MILCOM52596.2021.9653028

2. Sahingoz, O. K., Buber, E., Demir, O., & Diri, B. (2019). Machine learning based phishing detection from URLs. Expert Systems with Applications, 117, 345–357. https://doi.org/10.1016/j.eswa.2018.09.029

3. Do, N. Q., Selamat, A., Fujita, H., & Krejcar, O. (2024). An integrated model based on deep learning classifiers and pre-trained transformer for phishing URL detection. Future Generation Computer Systems, 161, 269–285. https://doi.org/10.1016/j.future.2024.06.031

4. Wang, C., et al. (2022). TCURL: Exploring hybrid transformer and convolutional neural network on phishing URL detection. Knowledge-Based Systems, 258, Article 109955. https://doi.org/10.1016/j.knosys.2022.109955

5. Rashid, F., Doyle, B., Han, S. C., & Seneviratne, S. (2024). Phishing URL detection generalisation using unsupervised domain adaptation. Computer Networks, 245, Article 110398. https://doi.org/10.1016/j.comnet.2024.110398

6. Haynes, K., et al. (2021). Lightweight URL-based phishing detection using natural language processing transformers for mobile devices. Procedia Computer Science, 191, 127–134. https://doi.org/10.1016/j.procs.2021.07.040

7. Aljofey, A., Jiang, Q., Qu, Q., Huang, M., & Niyigena, J. P. (2022). An effective phishing detection model based on character-level convolutional neural network from URL. Electronics, 11(9), 1514. https://doi.org/10.3390/electronics11091514

8. Le, H., Pham, Q., S time? (2018). URLNet: Learning a URL representation with deep learning for malicious URL detection. Proceedings of the 2018 IEEE Security and Privacy Workshops. https://doi.org/10.1109/SPW.2018.00026

9. Chiew, K. L., Tan, C. L., Wong, K. S., Yong, K. S. C., & Tiong, W. K. (2019). A new hybrid ensemble feature selection framework for machine learning-based phishing detection system. Information Sciences, 484, 153–166. https://doi.org/10.1016/j.ins.2019.01.064

10. Li, Y., et al. (2019). A stacking model using URL and HTML features for phishing webpage detection. Future Generation Computer Systems, 94, 27–39. https://doi.org/10.1016/j.future.2018.11.004

11. Bu, S., et al. (2021). Deep character-level anomaly detection based on a convolutional autoencoder for zero-day phishing URL detection. Electronics, 10(12), 1492. https://doi.org/10.3390/electronics10121492

12. Rasheed, A., et al. (2021). Adversarial attacks on featureless deep learning malicious URLs detection. Computers, Materials & Continua, 68(1), 921–939.

13. Al-Ahmadi, S., & AbuSharkh, M. (2022). PDGAN: Phishing detection with generative adversarial networks. IEEE Access, 10, 42459–42472. https://doi.org/10.1109/ACCESS.2022.3168235

14. Shirazi, H., et al. (2021). Towards performance of NLP transformers on URL-based phishing detection for mobile devices. International Journal of Ubiquitous Systems and Pervasive Networks.

15. Detecting phishing URLs using the BERT transformer model. (2023). 2023 IEEE International Conference on Big Data, 2483–2492. https://doi.org/10.1109/BigData59044.2023.10386782

Downloads

Published

25-01-2026

How to Cite

Detecting Phishing Websites Using Transformer-Based Language Models and URL Features. (2026). Innovative Journal of Computing, Security and Digital Technologies, 1(1), 23-28. https://ijcdst.in/index.php/about/article/view/9

Similar Articles

1-10 of 13

You may also start an advanced similarity search for this article.