An Adaptive Multi-Modal Deep Learning Approach for Real-Time Phishing URL Detection

Machine Learning algorithms are efficient to do binary classification and to detect the malicious URLs.URL detection using a machine learning model is that it accepts the URL as user input and detects and classifies it as benign or malicious one. Model does binary classification with 99% accuracy. This model can be used in the cyber security domain and to avoid digital attacks by knowing the malicious and benign URLs in prior. Safety measures can be taken if the URL is found malicious. In the future work, we would like to use different feature selection ensembles, clustering algorithms and feature engineering techniques for the hidden feature generation that helps in improving the detection accuracy of the model. Some of the future suggestions are: Need huge quantity of customized, structured training data.Needs rigorous training.To build ready-to-use machine learning models for detection of Malicious URLs.

[1]           P. Zhang, A. Oest, H. Cho, Z. Sun, R. Johnson, B. Wardman, S. Sarker, A. Kapravelos, T. Bao, R. Wang, Y. Shoshitaishvili, A. Doupe, and G.-J. Ahn, ‘‘Crawl Phish: Large-scale analysis of client-side cloaking techniques in phishing,’’ in Proc. IEEE Symp. Secur. Privacy (SP), May 2021, pp. 1109–1124. [2]           ENISA. (2021). Cybersecurity for SMEs—Challenges and Recommendations. [Online]. Available: https://www.enisa.europa.eu/publications/ enisa-report-cybersecurity-for-smes. [3]           Cisco. (2021). Cyber Security Threat Trends: Phishing, Crypto Top the List. [Online]. Available: https://umbrella.cisco.com/info/2021-cybersecurity-threat-trends-phishing-crypto-top-the-list. [4]           M. Alsharnouby, F. Alaca, and S. Chiasson, ‘‘Why phishing still works: User strategies for combating phishing attacks,’’ Int. J. Hum.-Comput. Stud., vol. 82, pp. 69–82, Oct. 2015. [5]           Anti-Phishing Working Group—APWG. (2022). Phishing Activity Trends Report-1Q. [Online]. Available: https://docs.apwg.org/reports/ apwg_trends_report_q1_2022.pdf. [6]           I. Vayansky and S. Kumar, ‘‘Phishing—Challenges and solutions,’’ Comput. Fraud Secur. vol. 2018, no. 1, pp. 15–20, Jan. 2018. [7]           Y. Cohen, D. Hendler, and A. Rubin, ‘‘Detection of malicious webmail attachments based on propagation patterns,’’ Knowl.-Based Syst., vol. 141, pp. 67–79, Feb. 2018. [8]           J. Yuan, Y. Liu, and L. Yu, ‘‘A novel approach for malicious URL detection based on the joint model,’’ Secur. Commun. Netw. vol. 2021, pp. 1–12, Dec. 2021. [9]           T. Li, G. Kou, and Y. Peng, ‘‘Improving malicious URLs detection via feature engineering: Linear and nonlinear space transformation methods,’’ Inf. Syst., vol. 91, Jul. 2020, Art. no. 101494. C. D. Xuan, H. Dinh, and T. Victor, ‘‘Malicious URL detection based on machine learning,’’ Int. J. Adv. Comput. Sci. Appl., vol. 11, no. 1, pp. 148–153, 2020.