Header

UZH-Logo

Maintenance Infos

How users tweet about a cyber attack: an explorative study using machine learning and social network analysis


Vogler, Daniel; Meissner, Florian (2020). How users tweet about a cyber attack: an explorative study using machine learning and social network analysis. Journal of digital media & policy, 11(2):195-214.

Abstract

Cybercrime is a growing threat for firms and customers that emerged with the digitization of business. However, research shows that even though people claim that they are concerned about their privacy online, they do not act correspondingly. This study investigates how prevalent security issues are during a cyber attack among Twitter users. The case under examination is the security breach at the US ticket sales company, Ticketfly, that compromised the information of 26 million users. Tweets related to cybersecurity are detected through the application of automated text classification based on supervised machine learning with support vector machines. Subsequently, the users that wrote security-related tweets are grouped into communities through a social network analysis. The results of this multi-method study show that users concerned about security issues are mostly part of expert communities with already superior knowledge about cybersecurity.

Abstract

Cybercrime is a growing threat for firms and customers that emerged with the digitization of business. However, research shows that even though people claim that they are concerned about their privacy online, they do not act correspondingly. This study investigates how prevalent security issues are during a cyber attack among Twitter users. The case under examination is the security breach at the US ticket sales company, Ticketfly, that compromised the information of 26 million users. Tweets related to cybersecurity are detected through the application of automated text classification based on supervised machine learning with support vector machines. Subsequently, the users that wrote security-related tweets are grouped into communities through a social network analysis. The results of this multi-method study show that users concerned about security issues are mostly part of expert communities with already superior knowledge about cybersecurity.

Statistics

Citations

Dimensions.ai Metrics
1 citation in Web of Science®
2 citations in Scopus®
Google Scholar™

Altmetrics

Additional indexing

Item Type:Journal Article, refereed, original work
Communities & Collections:06 Faculty of Arts > Department of Communication and Media Research
06 Faculty of Arts > Institute for Research on the Public Sphere and Society
Dewey Decimal Classification:070 News media, journalism & publishing
Uncontrolled Keywords:Twitter, cybercrime, cybersecurity awareness, data breach, machine learning, social network analysis, text classification
Language:English
Date:2020
Deposited On:29 Jan 2021 14:10
Last Modified:24 Jun 2024 01:43
Publisher:Intellect
ISSN:2516-3531
OA Status:Closed
Publisher DOI:https://doi.org/10.1386/jdmp_00016_1
Full text not available from this repository.