UPM Institutional Repository

An effective security alert mechanism for real-time phishing tweet detection on twitter


Mohd Sani, Nor Fazlida and Sharum, Mohd Yunus and Yaakob, Razali and Abdullah@Selimun, Mohd Taufik and Liew, Seow Wooi (2019) An effective security alert mechanism for real-time phishing tweet detection on twitter. Computers and Security, 83. pp. 201-207. ISSN 0167-4048


Phishing is a form of social engineering crime uses to deceive victims by directing them to a fraudulent website where their private and confidential information are collected for further illegal actions. Phishing attacks have now targeted users at Online Social Networks (OSN)s such as Twitter, Facebook, Myspace, etc. which traditionally, targeting email users. Twitter has become so prevalent to phishers to spread phishing attacks nowadays due to its vast information dissemination and difficult to be detected unlike email. As such, the effectiveness of security alert to prompt Twitter users for the tweet containing phishing Uniform Resource Locator (URL) in real-time is crucial. Many solutions have been proposed but their effectiveness are inadequate and doubtful. In this paper, we propose an effective security alert mechanism making use of a classification model derived from a supervised machine learning technique of Random Forest (RF) and the identified 11 best classification features yielded 94.75% accuracy higher than 94.56% yielded by other researchers who used more than 11 features trained on the same dataset collected from Twitter. To determine its effectiveness, we used 200 phishing URLs collected from Twitter and PhishTank respectively. From our experiment, we are able to justify that such proposed security alert mechanism managed to prompt 97.50% effectively the security alert to Twitter users in real-time.

Download File

[img] Text (Abstract)

Download (6kB)

Additional Metadata

Item Type: Article
Divisions: Faculty of Computer Science and Information Technology
DOI Number: https://doi.org/10.1016/j.cose.2019.02.004
Publisher: Elsevier
Keywords: Phishing; Twitter; Classification model; Random Forest (RF); Security alert mechanism
Depositing User: Ms. Nuraida Ibrahim
Date Deposited: 09 Nov 2020 15:28
Last Modified: 09 Nov 2020 15:28
Altmetrics: http://www.altmetric.com/details.php?domain=psasir.upm.edu.my&doi=10.1016/j.cose.2019.02.004
URI: http://psasir.upm.edu.my/id/eprint/80585
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item