Title: A hybrid model to detect phishing-websites

Authors: Mohamed Hafidi; Lamia Mahnane

Addresses: LRS Laboratory, Department of Computer Science, University of Annaba, Algeria ' LRS Laboratory, Department of Computer Science, University of Annaba, Algeria

Abstract: Phishing and fraud sites has been widespread on the internet in recent times, in which that it is becoming a source of great concern and a serious cyber security problem, as internet fraudsters target sensitive data and personal information of users, especially the username and password. Numerous approaches has been proposed and used in order to prevent and reduce these phishing websites and attacks, and protect users and their privacy. In this paper, we proposed an anti-phishing system based on machine learning algorithms and classifiers, combined with white-list and black-list techniques. We used the C4.5 classifier that generate a decision tree that provides highly accurate results in deciding whether a site is a fraudulent website or a genuine website based on its URL. Our predictive model achieved a high accuracy rate of 83%. We have modelled our system in the form of a desktop application so that the user can add the URL of the website needed to be checked, and the results appear to the user. The system combination accuracy can exceed 95%, and the execution results prove that.

Keywords: anti-phishing; machine learning; black-list; white-list; cyber security.

DOI: 10.1504/IJITST.2022.126472

International Journal of Internet Technology and Secured Transactions, 2022 Vol.12 No.6, pp.483 - 502

Received: 08 Jan 2021
Accepted: 24 Mar 2022

Published online: 26 Oct 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article