Proceedings of the E-Learning and Smart Engineering Systems (ELSES 2023)

URL_trigger: Real time solution for Detection Malicious URL using Deep Learning

Authors
Omar Lamrabti1, *, Abdellatif MezriOui1, Abdelhamid Belmekki1
1STRS LAB, RAISS TEAM, National Institute of Posts and Telecommunications, Rabat, Morocco
*Corresponding author. Email: o.lamrabti@inpt.ac.ma
Corresponding Author
Omar Lamrabti
Available Online 5 February 2024.
DOI
10.2991/978-94-6463-360-3_33How to use a DOI?
Keywords
Deep Learning; Machine Learning; Malicious URL; URL_trigger
Abstract

Evolution of technology has many drawbacks as well as benefits, requiring a great investment to ensure security where the human component is the core of this equation. To that end, malicious URLs are among tactics that target users unawareness, unable to distinguish between legitimate and malicious URLs, inducing them to interact with malicious links that hosts varieties of malicious contents such as malware, phishing, or drive-by downloads to carry out cyberattacks. To contribute to this defiance, research focused on Machine Learning (ML) systems. Unfortunately, to withstand new attacks via ML, features collection must be a continuous task that consumes time and energy, which is considered a major drawback. However, Deep Learning (DL) mitigates this defiance by learning from unstructured data without supervision. Nevertheless, adoption of DL doesn’t cover all well-known DL models in one experience. In this paper, we introduce “URL_trigger”, a solution based on DL and ML in order to detect malicious URLs. To reach this goal, the solution includes an intelligent system that collects and detects malicious URLs from various sources (e.g. twitter, zone-h, pastebin, virustotal) that will be saved continuously in our dataset. URL_trigger provides real-time evaluation of malicious URLs via various techniques, including: blacklisting, lexical features, host-based features, content-based features, machine learning, and deep learning. We use CNN, RNN, RCNN, and DNN as models for learning URL_trigger. Achieving an accuracy of 99% confirms the effectiveness of our system in terms of detecting malicious links compared to other solutions that use DL.

Copyright
© 2023 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

Volume Title
Proceedings of the E-Learning and Smart Engineering Systems (ELSES 2023)
Series
Atlantis Highlights in Social Sciences, Education and Humanities
Publication Date
5 February 2024
ISBN
978-94-6463-360-3
ISSN
2667-128X
DOI
10.2991/978-94-6463-360-3_33How to use a DOI?
Copyright
© 2023 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Cite this article

TY  - CONF
AU  - Omar Lamrabti
AU  - Abdellatif MezriOui
AU  - Abdelhamid Belmekki
PY  - 2024
DA  - 2024/02/05
TI  - URL_trigger: Real time solution for Detection Malicious URL using Deep Learning
BT  - Proceedings of the E-Learning and Smart Engineering Systems (ELSES 2023)
PB  - Atlantis Press
SP  - 328
EP  - 334
SN  - 2667-128X
UR  - https://doi.org/10.2991/978-94-6463-360-3_33
DO  - 10.2991/978-94-6463-360-3_33
ID  - Lamrabti2024
ER  -