The Journal of Web Science > Vol 4 > Issue 3

Predicting Online Islamophobic Behavior after #ParisAttacks

Kareem Darwish, Qatar Computing Research Institute, Hamad bin Khalifa University, Qatar, Walid Magdy, School of Informatics, The University of Edinburgh, UK, Afshin Rahimi, The University of Melbourne, Australia, Timothy Baldwin, The University of Melbourne, Australia, Norah Abokhodair, Microsoft, USA,
Suggested Citation
Kareem Darwish, Walid Magdy, Afshin Rahimi, Timothy Baldwin and Norah Abokhodair (2018), "Predicting Online Islamophobic Behavior after #ParisAttacks", The Journal of Web Science: Vol. 4: No. 3, pp 34-52.

Publication Date: 19 Mar 2018
© 2018 K. Darwish, W. Magdy, A. Rahimi, T. Baldwin and N. Abokhodair
IslamophobiaNetwork AnalysisTwitterHomophilySocial NetworksParis attacksTerrorist AttacksStance Prediction


Open Access

This is published under the terms of CC BY-NC-ND 2.0.

In this article:
1. Introduction
2. Background
3. Post-Attack Data Collection
4. Statistics on the Data
5. Pre-Attack Prediction
6. Discussion
7. Conclusion
A. Top Features


The tragic Paris terrorist attacks of November 13, 2015 sparked a massive global discussion on Twitter and other social media, with millions of tweets in the first few hours after the attacks. Most of these tweets were condemning the attacks and showing support for Parisians. One of the trending debates related to the attacks concerned possible association between Muslims and terrorism, which resulted in a world-wide debate between those attacking and those defending Islam. In this paper, we use this incident as a case study to examine using online social network interactions prior to an event to predict what attitudes will be expressed in response to the event. Specifically, we focus on how a person’s online content and network dynamics can be used to predict future attitudes and stance in the aftermath of a major event. In our study, we collected a set of 8.36 million tweets related to the Paris attacks within the 50 hours following the event, of which we identified over 900k tweets mentioning Islam and Muslims. We then quantitatively analyzed users’ network interactions and historical tweets to predict their attitudes towards Islam and Muslims. We provide a description of the quantitative results based on the tweet content (hashtags) and network interactions (retweets, replies, and mentions). We analyze two types of data: (1) we use post-event tweets to learn users’ stated stance towards Muslims based on sampling methods and crowd-sourced annotations; and (2) we employ pre-event interactions on Twitter to build a classifier to predict post-event stance. We found that pre-event network interactions can predict attitudes towards Muslims with 82% macro F-measure, even in the absence of prior mentions of Islam, Muslims, or related terms.