Shanthosh. S, Dr. G. Rosline Nesa kumari

DOI: https://doi.org/10.5281/zenodo.23732245

The World Wide Web has grown to be a vital component of our daily lives for the exchange of information and the sharing of knowledge. It facilitates timely, quick, and simple information transactions. Theft of identification and fraud involving identity are two forms of cybercrime in which hackers and malicious users take the personal data of authorized users in an attempt to perpetrate fraud or deception for financial gain. Because rogue URLs host unwanted material (email spam, scams, brute force flaws, etc.) to trick gullible people into falling for schemes (money loss, private information theft, malicious software deployment, etc.). Systems should be quick, accurate, and able to identify new malicious content in order to detect such crimes. Traditionally, blacklists are used primarily for this detection. Blacklists, however, are not all-inclusive and are unable to identify recently created dangerous URLs. In an effort to increase the breadth of malicious URL detectors, machine learning techniques have been the focus of more study in recent years. In order to determine whether a URL is good or poor, employ a straightforward technique in this work, which contrasts with SVM and LR, two additional methods.

Detection of malicious URLs, machine learning, CNN, SVM, Logistic Regression, Cybersecurity, Phishing Detection, URL Threat Classification.

For many cybersecurity applications, malicious URL detection is essential, and deep learning techniques are undoubtedly a viable avenue. This work discusses the CNN methodology focused on the word2vac feature, logistic regression, as well as the support vector algorithm strategy based on term prevalence–inverse file rate. The three characteristics of SVM, logical regression, and CNN—precision, recall, and fl-socre—can be compared to draw a conclusion. The following three column tables demonstrate how the logical regression method uses the term frequency–inverse document frequency of SVM. The precision, recall, and fl-socre of SVM are all marginally greater than those of the logical regression methodology. Based on the inverse document frequency and term frequency, the SVM approach is compatible with the Word2vac-based convolution neural network.

[1]approach for detecting malicious javascript code,” Security and Communication Networks, 2016.
[2]]Anjali B. Sayamber ,Arati M. Dixit , “Malicious URL Detection and Identification”, International Journal of Computer Applications (0975 – 8887) Volume 99 – No.17, August 2014.
[3]D. R. Patil and J. Patil, “Survey on malicious web pages detection techniques,” International Journal of u-and e-Service, Science and Technology, vol. 8, no. 5, pp. 195–206, 2015.
[4]Usha Narra, Corrado Aaron Visaggio, Mark Stamp, Thomas H. Austin, “Clustering versus SVM for malware detection”, Springer, Journal of Computer Virology and Hacking Techniques 10/2015.
[5]R. K. Nepali and Y. Wang, “You look suspicious!!: Leveraging visible attributes to classify malicious short urls on twitter,” in 2016 49th Hawaii International Conference on System Sciences (HICSS). IEEE, 2016, pp. 2648–2655.
[6]M.D.Zeiler,“Adadelta:anadaptivelearningratemethod,”arXivpreprintarXiv:1212.5701, 2012.
[7]L. Xu, Z. Zhan, S. Xu, and K. Ye, “Cross-layer detection of malicious websites,” in Proceedings of the third ACM conference on Data and application security and privacy. ACM, 2013, pp. 141–152.
[8]B.-I. Kim, C.-T. Im, and H.-C. Jung, “Suspicious malicious web site detection with strength analysis of a javascript obfuscation,” Interna-tional Journal of Advanced Science and Technology, vol. 26, pp. 19–32, 2011.
[9]E. Sorio, A. Bartoli, and E. Medvet, “Detection of hidden fraudulent urls within trusted sites using lexical features,” in Availability, Relia-bility and Security (ARES), 2013 Eighth International Conference on. IEEE, 2013, pp. 242–247.
[10]Liu Aijiang,HuangChanghui and Hu Guangjun,”Detection Method of Trojan’s Control Domain Based on Improved Neural Network Algorithm,”China Academic Journal Electronic Publishing House,2014.
[11]SHA Hong-zhou,ZHOUZhou,LIU Qing-yun and QIN Peng,“Light-weight self-learning approach for URL classification”,Journal on Communications,2014.
[12]YUAN Fu-xiang ,LIU Fen-lin ,LU Bin and GONG Dao-fu ,“Anomaly domains detection algorithm based on historical data”Journal on Communications,2016.
[13]Doyen Sahoo, Chenghao Liu, and Steven C.H. Hoi,”Malicious URL Detection using Machine Learning: A Survey”,2017.
[14]S. Sheng, B. Wardman, G. Warner, L. F. Cranor, J. Hong, and C. Zhang, “An empirical analysis of phishing blacklists,” in Proceedings of Sixth Conference on Email and Anti-Spam (CEAS), 2009.
[15]S. Sinha, M. Bailey, and F. Jahanian, “Shades of grey: On the effectiveness of reputation-based “blacklists”,” in Malicious and Unwanted Software, 2008. MALWARE 2008. 3rd International Conference on. IEEE, 2008, pp. 57–64.
[16]A. Budaraju and J. D. R. Pesaramilli, “Intelligent Threat Analysis and Fault Resilience Mechanism for Cloud Banking Microservices,” in Innovative Computing and Communications, ICICC 2026, Lecture Notes in Networks and Systems, vol. 2037. Springer, Cham, 2026. doi: [10.1007/978-3-032-30008-9_7]
[17]J. Pesaramilli and T. Gudisa, “Real-Time Attack Surface Reduction in Cloud Infrastructures Using Reinforcement Learning-Driven Cyber Deception Strategies,” 2025. doi: [10.1109/ICONSTEM65670.2025.11374717]
[18]J. David and A. Budaraju, “Self-Healing Microservices for Cloud Banking with Anomaly-Driven Threat Detection and Fault Tolerance,” in Proceedings of the International Conference on Computing Systems and Intelligent Applications (ComSIA 2026), Lecture Notes in Networks and Systems, vol. 2054, Springer, 2026, pp. 524–541.
[19]A. Kumar Jonnalagadda, B. Madupati, R. Varma Vegesna, and S. Kumar Vududala, “Federated Data Modeling for LLM Deployment in Secure Cloud-Native Architectures,” in 2025 International Conference on Computing Technologies & Data Communication (ICCTDC), Hassan, India, 2025, pp. 1–8, doi: [10.1109/ICCTDC64446.2025.11158782]
[20]B. Madupati, A. K. Jonnalagadda, S. K. Vududala, and R. V. Vegesna, “Blockchain and AI Fusion for Financial Traceability and Cyber-Resilience in Industry 5.0 Environments,” in 2026 Innovations in Machine, Engineering, and Digital Conference (IMED), Kota Kinabalu, Malaysia, 2026, pp. 1–7, doi: [10.1109/IMED68921.2026.1148499]

Shanthosh. S, Dr. G. Rosline Nesa kumari (2026). Hybrid Machine Learning and Deep Learning Model for Malicious URL Detection. International Journal of Engineering and Techniques, 12(5), 395–399. ISSN: 2395-1303. DOI: https://doi.org/10.5281/zenodo.23732245