GAX and GAS Modes: A New Hybrid Deep Learning Method for Phishing URL Detection using GRU, Attention Mechanisms, SVM and XGBOOST Algorithms

Hajara Musa, M. S. Adamu, A. Y. Gital, Usman Ali, A. M. Kwami, F. U. Zambuk, A. A. Aminu

Abstract


Many approaches have been proposed for classifying and detecting of phishing URLs. While these methods have recorded great successes, many research are based on content based approaches which has poor generalization ability against new unseen URLS and it requiring domain knowledge experts and substantial future engineering. To cover these limitations, recently, our approaches which offer the ability to automatically capture the semantic or sequential patterns in URLs have been proposed by several studies. This paper reviews and compares the state-of-the-art approaches for phishing URLs classification. It has been observed from the experimental results on 131,072 URLs dataset that Gated Recurrent Unit–attention mechanism- support vector machine (GRU-ATT-SVM) model outperforms the basic from other in terms of accuracy, precision, recall, training, and testing time. Hence the GRU-ATT-SVM model can be the best for the phishing URL classification in the future work.  


Full Text:

PDF

References


Musa. H., Ya, A., Alkali, A. H., & Umar, A. (2022). ). “Phishing URLs Classification Using

Deep Learning Approach”. Journal of Science Technology and Education 10(2), 413–424.

Yadav, D. P., Paliwal, P., KumaD, D., and Tripathi, R. (2017). A Novel Ensemble Based

Identification of Phishing E-Mails, 2–6.

Yang, P., Zhao, G., & Zeng, P. (2019). Phishing website detection based on multidimensional features driven by deep learning. IEEE Access, 7. https://doi.org/10.1109/ACCESS.2019.2892066

Ahmad A. Y, Selvakumar M., Mohammed A., Mohammed A. and Samer A. S. (2016). TrustQR: A New Technique for the Detection of Phishing Attacks on QR Code, Adv. Sci. Lett., vol. 22, no. 10, pp. 2905-2909, Oct. 2016.

Inez C. C. and Baruch F. (2018). Setting Priorities in Behavioral Interventions: An Application to Reducing Phishing Risk, Risk Anal., vol. 38, no. 4, pp. 826-838, Apr. 2018.

Musa, H., Gital, A. Y., Zambuk, F. U., Umar, A., Umar, A. Y., & Waziri, J. U. (2019). A comparative analysis of phishing website detection using XGBOOST algorithm. Journal of Theoretical and Applied Information Technology, 97(5).

A. Ali, Q. Jiang, Q. Qu, M. Huang, and J. Niyigena, "An effective phishing detection model based on character level convolutional neural network from URL," Electronics 9, no. 9 (2020): 1514.

L. Yann, Y. Bengio, and G. Hinton. "Deep learning." nature 521, no. 7553 (2015): 436-444.

C. Kyunghyun, B V. Merriënboer, C. Gulcehre, D. Bahdanau, F. Bougares, H. Schwenk, and Y. Bengio, "Learning phrase representations using RNN encoder-decoder for statistical machine translation," arXiv preprint arXiv:1406.1078 (2014)

B. Yun, J. Xie, C. Liu, Y.Tao, B. Zeng, and C. Li, "Regression modeling for enterprise electricity consumption: A comparison of recurrent neural network and its variants," International Journal of Electrical Power & Energy Systems 126 (2021): 106612.

Wu, T., Wang, M., Xi, Y., Zhao, Z. (2022): Malicious URL Detection Model Based on Bidirectional Gated Recurrent Unit and Attention Mechanism. Appl. Sci. 2022, 12, 12367. https://doi.org/10.3390/ app122312367

Tianqi Chen Oct. 22 2014. Introduction to Boosted Trees, university of Washington.

brahim, A., Kashef, R., Li, M., Valencia, E., & Huang, E. (2020). Bitcoin Network

Mechanics : Forecasting the BTC Closing Price Using Vector Auto-Regression Models Based on Endogenous and Exogenous Feature Variables.J. Risk Financial Manag., MDPI, 13, 189; doi:10.3390/jrfm13090189.

Cohen, G. (2020). Forecasting Bitcoin Trends Using Algorithmic Learning Systems Learning Systems, Entropy J., MDPI vol. i, 2020.

Brownlee, J. (2019). Master Machine Learning Algorithms. Discover How they Work and Implement Them from Scratch. Edition, v1.1 http://MachineLearningMastery.com.

R. Tomas, and L. Dovydaitis, "Detection of phishing URLs by using deep learning approach and multiple features combinations," Baltic journal of modern computing 8, no. 3 (2020): 471-483.

B. A. Correa, E. C. Bohorquez, S. Villegas, J. Vargas, and F. A. González, "Classifying phishing URLs using recurrent neural networks," In 2017 APWG symposium on electronic crime research (eCrime), pp. 1-8. IEEE, 2017.

Musa, H., Aminu, A. A., & Muhammad, A. N. (2019). Investigation the Effect of Dataset Size on the Performance of Different algorithm in Phishing Website Detection, International Journal of Artificial Intelligence and Applications (IJAIA), Vol.10, No.3, May 2019 pp, 53–59. https://doi.org/10.9790/7388-0901025359

Musa, H., Gital, A. Y., Bitrus, M. G., Juma, F., & Balde, M. A. (2020). Boosting the Accuracy of Phishing Detection with Less Features Using XGBOOST. 8(2), 81–90.

Musa, H., Gital, A. Y., Zambuk, F. U., Umar, A., Umar, A. Y., & Waziri, J. U. (2019). A comparative analysis of phishing website detection using XGBOOST algorithm. Journal of Theoretical and Applied Information Technology, 97(5).

Musa H., Gital, A. Y., A, U. A., & Kwami, A. M. (2023). A comprehensive Review on phishing website detection using machine learning and deep learning techniques. 8(2), 188–198.


Refbacks

  • There are currently no refbacks.