ChapterPDF Available

Abstract and Figures

Social media are widely used by terrorist organizations and extremist groups for disseminating propaganda and recruiting new members. Given the recent pledges both by the major social media platforms and governments towards combating online terrorism, our work aims at understanding the terrorism-related content posted on social media and distinguishing accounts of relevance to terrorism investigations from innocuous ones. We conducted an analysis of textual, spatial, temporal and social network features on data and metadata gathered from suspended Twitter content, and compared them with non-suspended content. Our analysis reveals a number of distinct characteristics of terrorism-related Twitter accounts. This work is a first step towards automated tools for the early detection of terrorism-related and extremist content on Twitter.
Content may be subject to copyright.
Chapter 11
Analysis of Suspended Terrorism-Related
Content on Social Media
George Kalpakis, Theodora Tsikrika, Ilias Gialampoukidis,
Symeon Papadopoulos, Stefanos Vrochidis, and Ioannis Kompatsiaris
Several popular social media platforms that emerged during the past decade
have revolutionized modern communications among people worldwide cutting
across different nationalities, cultures, and residences, and have resulted in the
development of online communities providing the means for the open sharing of
information. However, due to their broad reach, social media are also being used
with subversive intentions. For instance, in recent years they have been employed
by terrorist and extremist groups for further supporting their goals of spreading
their propaganda, radicalizing new members, and disseminating material targeting
potential perpetrators of future attacks. Therefore, online social networks present a
digital space of particular interest to governments, law enforcement agencies and
social media companies in their effort to suppress terrorism content.
Identifying terrorism-related content in social media is a challenging task.
Social media platforms host overwhelming amounts of discussions, posted daily by
millions of people, making it practically impossible to detect extremist or terrorism-
related content by solely relying on manual inspection performed by content
moderators. Therefore, an interesting research question is whether the analysis
of social media content can result in identifying multiple complementary weak
G. Kalpakis · T. Tsikrika · I. Gialampoukidis · S. Papadopoulos · S. Vrochidis ()
I. Kompatsiaris
Information Technologies Institute, Centre for Research and Technology Hellas
Thermi-Thessaloniki, Thermi, Greece
© The Author(s) 2018
G. Leventakis, M. R. Haberfeld (eds.), Community-Oriented Policing
and Technological Innovations, SpringerBriefs in Criminology, 8_11
108 G. Kalpakis et al.
signals revealing the distinctive nature of terrorism-related content, and thus be an
additional means towards the automatic or semi-automatic detection and subsequent
removal of such content.
In this context, this work aims at analyzing the particular traits of terrorism-
related content published on Twitter, a popular channel among terrorist groups, with
the goal to distinguish terrorism-related accounts from others. To this end, a dataset
of terrorism-related content was collected from Twitter through searches based on
terrorism-related keywords provided by domain experts. In our study, we analyzed
several textual, spatial, temporal and social network features of the gathered posts
and their metadata and compared them against “neutral” Twitter content.
Our study unveiled a number of distinct characteristics of extremism and
terrorism-related Twitter accounts and paves the path towards the development
of automated tools that aim at leveraging the distinct traits of terrorism-related
accounts for the early detection of terrorist and extremist content in Twitter, with
the goal of alerting social media companies about the presence of such posts and
speeding up the process of removing them from their platforms. This work is
particularly timely given the recent pledges both by major social media platforms
and governments around the world to step up their efforts towards countering online
abusive content (Twitter Public Policy 2017).
Section “Related Work” summarizes related work in the area. Section “Data
Collection and Analysis” describes the methodology and the dataset collected
for our analysis. Section “Experimental Results” presents the findings of our
analysis. Finally, Section “Conclusions” concludes with an outline of future research
Related Work
Related research conducted in the past years has focused on examining the nature
of terrorism-related content published by participants in extremist Web forums.
Specifically, several research efforts have proposed methods for analyzing extremist
Web forums for detecting users representing potential lone wolf terrorists and
perpetrators of radical violence (Johansson et al. 2013; Scanlon and Gerber 2014).
Additionally, the use of social media by terrorist and extremist groups and the
resulting social network perspectives have also been studied. Recent works have
examined the use of social media platforms by terrorist groups and organizations
(Chatfield et al. 2015; Klausen 2015). Moreover, key player and key community
identification in terrorism-related Twitter networks has been addressed through
the use of different centrality measures and community detection algorithms
(Gialampoukidis et al. 2016,2017). Complementary to the aforementioned research
11 Analysis of Suspended Terrorism-Related Content on Social Media 109
efforts, our paper analyzes several textual, spatial, temporal and social network
features which, when combined, are capable of characterizing the terrorism-related
nature of Twitter accounts.
Data Collection and Analysis
Our investigation focused on the Twitter platform given its popularity among
terrorist groups as a means for spreading their propaganda and recruiting new
members (Klausen 2015). Under the pressure put during the recent years by
governments around the world to combat online extremism, Twitter has made
significant efforts towards blocking accounts that promote terrorism and violence
(Twitter Inc. 2016). In particular, Twitter has been suspending user accounts based
on whether they are exhibiting abusive behavior that violates its rules,1including
posting content related to violent threats, hate speech, and terrorism. To this end,
Twitter has suspended 636,000 accounts between August 2015 and December 2016,
with more than half of them occurring in the last 6 months of 2016 (Larson 2017).
In this context, we consider that Twitter accounts that have been posting
content related to terrorism and are suspended at some point in time represent in
principle users who have been exploiting the social media platform for serving
their subversive intentions and promoting terrorism in general, e.g., disseminating
propaganda, etc. Twitter accounts that have been posting content related to terrorism
and have not been suspended are generally considered as users interested in the
domain (e.g., posting news related to terrorism attacks), but without subversive
intentions. There may of course be cases where non-suspended users also have
darker motives and are actively engaging in propaganda and radicalization efforts,
but have not thus far been detected so that they can be suspended by Twitter. In this
work, we consider that this phenomenon might occur indeed but that it is less likely
given Twitter’s efforts in this direction.
Our analysis is based on the comparison of various characteristics of suspended
Twitter accounts against those of non-suspended accounts. Both types of account
post content relevant to the terrorism domain. The goal of our study is to determine
the key factors that are capable of providing weak signals for distinguishing among
ordinary Twitter users and those with subversive behavior based on the analysis of a
variety of textual, spatial, temporal and social network features. The comparison
is performed by examining the lifetime of suspended accounts, analyzing user
accounts from the social network perspective (i.e. based on their connectivity with
other user accounts), and exploiting geolocation information extracted from the
textual content of user posts.
110 G. Kalpakis et al.
Data Collection
The data for our study were collected using a social media crawling tool (Schinas
et al. 2017) capable of running queries on the Twitter API2based on a set of five
Arabic keywords related to terrorism propaganda. These keywords were provided
by law enforcement agents and domain experts in the context of the activities of
the EC-funded H2020 TENSOR3project and are related to the Caliphate, its news,
publications and photos from the Caliphate area.
The crawling tool ran for a 7-month period, and specifically from February 9 to
September 8, 2017, collecting tweets relevant to the provided keywords, along with
information about the user accounts that published this content. Our dataset consists
of 60,519 tweets posted by 33,827 Twitter users, with 4,967 accounts (14.70%)
having been suspended by Twitter within this period.
For each tweet in our dataset, we stored its textual content together with relevant
metadata, such as its URL address, the language used, its creation date, and the
number of likes, shares, comments, and views. Similarly, for each user account
having posted at least one tweet within our collection, we have captured its name,
username, and creation date along with the number of its friends, followers, items
(i.e. the total number of posts), favorites, and public lists that they are a member of.
Additionally, each user account was monitored on a daily basis to determine
whether it has been suspended by Twitter. Given that Twitter does not provide
information regarding the exact suspension date and time, this was determined based
on the latest post published by a suspended account. Finally, after processing the
data gathered, we built a social network graph representing the connectivity among
Twitter accounts based on user mentions.
Experimental Results
This section presents the findings of our comparison between the suspended and
non-suspended Twitter accounts on our dataset.
User Account Lifetime
First, we discuss the suspended Twitter account lifetime (see Fig. 11.1). Their
lifetime is determined by computing the difference between the suspension date
and the creation date of an account. The majority of suspended accounts (61.26%)
11 Analysis of Suspended Terrorism-Related Content on Social Media 111
Fig. 11.1 Lifetime of suspended accounts
have very short lifetime, fluctuating between 1 and 3 days, which is explained by
the efforts put by Twitter towards removing extremist content the moment it is
posted. However, an interesting finding is that a significant portion of the suspended
accounts (25.35%) have a lifetime longer than 30 days, which indicates that some
accounts manage to evade the monitoring processes of Twitter for longer periods.
Analysis of Mention Networks
The analysis of mention networks formed by users in our dataset provides insights to
the comparison between suspended and non-suspended accounts. The connectivity
for the two account types differs with respect to the interconnection of accounts
of the same type. In particular, 42.22% of the suspended accounts mention other
suspended users, whereas 52.66% mention non-suspended accounts (the remaining
5.12% of the suspended accounts mention users are not included in our dataset and
hence their suspension status is unknown). On the contrary, only 2.67% of non-
suspended accounts mention suspended users, whereas 89.91% are connected with
non-suspended users; again, the remaining 7.42% mention users not included in the
dataset. This behavior reveals a community-like behavior, where accounts of the
same type work together to fulfill their goals.
The connectivity pattern observed on the mention network is illustrated in the
suspended to non-suspended mention ratio plot (see Fig. 11.2). The peak observed
for mention ratio values fluctuating between 1 and 1.5 in the graph referring to
suspended accounts indicates that a significant part of the them is connected to a
larger number of suspended than non-suspended accounts, despite the fact that the
vast majority of accounts gathered in our dataset are non-suspended users.
112 G. Kalpakis et al.
Fig. 11.2 Suspended to
non-suspended mention ratio
Fig. 11.3 Twitter account
Friends and Followers
Figures 11.3 and 11.4 illustrate the distribution of numbers of friends and followers
per account type, respectively. In both cases, the vast majority of suspended users
have less than 100 friends or followers, which comes in contrast with the connec-
tivity of non-suspended accounts. The short lifetime of terrorism-related accounts
(due to their suspension by Twitter) could be a determining factor regarding their
number of connections.
11 Analysis of Suspended Terrorism-Related Content on Social Media 113
Fig. 11.4 Twitter account
Fig. 11.5 Twitter account
Posts, Favorites, and Lists
A similar trend is observed regarding the number of posted items and favorites
per account type (see Figs. 11.5 and 11.6, respectively). The number of posts
and favorites for the majority of suspended users is less than 100, whereas non-
suspended accounts exhibit the inverse behavior.
114 G. Kalpakis et al.
Fig. 11.6 Twitter account
Fig. 11.7 Twitter account
post rate
On the other hand, a different behavior is observed regarding the post rate (i.e. the
number of posts per day) per account type (see Fig. 11.7). The majority of suspended
accounts exhibit a post rate between 11 and 100 posts per day, whereas more than
half of ordinary Twitter accounts post less than 10 tweets per day. This indicates that
11 Analysis of Suspended Terrorism-Related Content on Social Media 115
Fig. 11.8 Number of times
during their short lifetime, suspended accounts tend to post a relatively large number
of tweets, possibly in an effort to disseminate many different pieces of information
for spreading their propaganda.
Significant differences are also observed with respect to the number of public
lists an account is a member of (see Fig. 11.8). The vast majority of suspended
accounts (71.89%) are not a member in any list, whereas more than half of the non-
suspended users are part of at least one list (54.48%), with almost one quarter of
ordinary Twitter accounts being members in more than five lists.
Spatial Distribution of Accounts
To delve into the spatial distribution of accounts, we performed text-based analysis
of the textual content of Twitter posts. We inferred the location of posts, even
in cases when it was not explicitly available through the geotagging metadata
accompanying a tweet. Geolocation inference from text was based on the approach
by Kordopatis-Zilos et al. (2017), which employs refined language models learned
from massive corpora of social media annotations. The results of the geolocation
extraction for the posts of suspended and non-suspended users are presented in Figs.
11.9 and 11.10 respectively. Given that our dataset is retrieved based on a set of
Arabic keywords, the geolocation information extracted for posts produced by both
account types refers to countries from the Middle East and Northern Africa, whereas
posts coming either from the United Arab Emirates or Syria are mostly associated
with suspended accounts.
116 G. Kalpakis et al.
Fig. 11.9 Inferred locations from posts by suspended Twitter accounts
Fig. 11.10 Inferred locations from posts by non-suspended Twitter accounts
11 Analysis of Suspended Terrorism-Related Content on Social Media 117
This paper aimed at understanding terrorism-related content on social media given
their increasing employment by terrorist organizations for spreading their propa-
ganda. We conducted an analysis on terrorism-related content posted on Twitter
focusing on the differences between suspended and non-suspended accounts. Our
analysis suggests that the traits observed in suspended users are different from
non-suspended ones from several different perspectives, namely textual, spatial,
temporal and social network features. These findings have the potential to set the
basis for automated methods that detect accounts that are likely associated with
abusive terrorism-related behavior. To this end, future work includes a more in-depth
and large-scale analysis of features presented here, as well as taking into account
additional features, including multimedia content such as images and videos.
Acknowledgements This work was supported by the TENSOR project (H2020-700024), funded
by the European Commission.
Chatfield, A. T., Reddick, C. G., & Brajawidagda, U. (2015). Tweeting propaganda, radicalization
and recruitment: Islamic state supporters multi-sided twitter networks. In Proceedings of the
16th Annual International Conference on Digital Government Research (pp. 239–249).
Gialampoukidis, I., Kalpakis, G., Tsikrika, T., Vrochidis, S., & Kompatsiaris, I. (2016). Key
player identification in terrorism-related social media networks using centrality measures. In
Intelligence and Security Informatics Conference (EISIC), 2016 European (pp. 112–115).
Gialampoukidis, I., Kalpakis, G., Tsikrika, T., Papadopoulos, S., Vrochidis, S., & Kompatsiaris,
I. (2017). Detection of terrorism-related twitter communities using centrality scores. In
Proceedings of the 2nd International Workshop on Multimedia Forensics and Security (pp.
Johansson, F., Kaati, L., & Shrestha, A. (2013). Detecting multiple aliases in social media. In
Proceedings of the 2013 IEEE/ACM International Conference on Advances in Social Networks
Analysis and Mining (pp. 1004–1011).
Klausen, J. (2015). Tweeting the Jihad: Social media networks of western foreign fighters in Syria
and Iraq. Studies in Conflict & Terrorism, 38(1), 1–22.
Kordopatis-Zilos, G., Papadopoulos, S., & Kompatsiaris, I. (2017). Geotagging text content with
language models and feature mining. In Proceedings of the IEEE.
Larson S. (2017). Twitter suspends 377,000 accounts for pro-terrorism content.http:// Accessed
15 Sept 2017.
Scanlon, J. R., & Gerber, M. S. (2014). Automatic detection of cyber-recruitment by violent
extremists. Security Informatics, 3(1), 5.
118 G. Kalpakis et al.
Schinas, M., Papadopoulos, S., Apostolidis, L., Kompatsiaris, Y., & Pericles, M. (2017). Open-
source monitoring, search and analytics over social media. In Proceedings of Internet Science
Conference. Springer.
Twitter Inc. (2016). Combating violent extremism.
combating-violent-extremism.html. Accessed 15 Sept 2017.
Twitter Public Policy. (2017). Global internet forum to counter terrorism.
Accessed 15 Sept 2017.
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0
International License (, which permits 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.
The images or other third party material in this chapter are included in the chapter’s Creative
Commons license, unless indicated otherwise in a credit line to the material. If material is not
included in the chapter’s Creative Commons license and your intended use is not permitted by
statutory regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder.
... Although, social media platforms have policies to avoid posting such content and have constantly been refining them (Kalpakis et al., 2018). But the terrorists have been circumventing all the guidelines and are persistently maintaining their followers through these platforms. ...
Full-text available
In this contemporary era, where a large part of the world population is deluged by extensive use of the internet and social media, terrorists have found it a potential opportunity to execute their vicious plans. They have got a befitting medium to reach out to their targets to spread propaganda, disseminate training content, operate virtually, and further their goals. To restrain such activities, information over the internet in context of terrorism needs to be analyzed to channel it to appropriate measures in combating terrorism. Open Source Intelligence (OSINT) accounts for a felicitous solution to this problem, which is an emerging discipline of leveraging publicly accessible sources of information over the internet by effectively utilizing it to extract intelligence. The process of OSINT extraction is broadly observed to be in three phases (i) Data Acquisition, (ii) Data Enrichment, and (iii) Knowledge Inference. In the context of terrorism, researchers have given noticeable contributions in compliance with these three phases. However, a comprehensive review that delineates these research contributions into an integrated workflow of intelligence extraction has not been found. The paper presents the most current review in OSINT, reflecting how the various state‐of‐the‐art tools and techniques can be applied in extracting terrorism‐related textual information from publicly accessible sources. Various data mining and text analysis‐based techniques, that is, natural language processing, machine learning, and deep learning have been reviewed to extract and evaluate textual data. Additionally, towards the end of the paper, we discuss challenges and gaps observed in different phases of OSINT extraction. This article is categorized under: Application Areas > Government and Public Sector Commercial, Legal, and Ethical Issues > Social Considerations Fundamental Concepts of Data and Knowledge > Motivation and Emergence of Data Mining
... The idea of investigating online activities and their ascribed meaning emerged from social sciences literature, which argues that an individual's belief systems, values, motivations, etc. can be inferred from their online activities (Golbeck et al., 2011), and are correlated with their real-world behaviours (Kosinski et al., 2013). Recognising the feasibility of identifying and assessing terrorist online activities, there has been a growing trend where researchers have harnessed the online information collected from terrorists and subjected it to empirical inquiry (Cohen et al., 2014;Kalpakis et al., 2018). For example, Pressman and Ivan (2016) In short, many of the online terrorists' supporters are not what Taylor and Horgan (2006) would describe as full-fledged terrorists, but this, of course, does not dictate that they would not become terrorists in the future. ...
Full-text available
The ubiquity of online radicalisation warrants the need for law enforcement to learn how to harness open-source intelligence to assess the threat posed and identify potential terrorists in advance. Although much is known about the online radicalisation process, there is a lack of risk assessment instruments developed specifically for online radicalisation, as well as little empirical research in this field. To address these research gaps, this thesis aims to articulate a set of person-centric and psychosocial risk factors and protective factors for identifying radicalisation in social media posts. Five studies were conducted. The first study (chapter 3) involved the use of a modified Delphi technique and literature review to derive a list of 12 factors and 42 indicators. The second study (chapter 4) served as a pilot to map the said factors and indicators to two groups of supporters of terrorists who have been influenced by radical online propaganda. These two groups comprise individuals who travelled to join the Islamic State in Iraq and Syria [ISIS] (i.e., foreign fighters), and individuals who did not travel to join ISIS (i.e., sympathisers). Next, the third study (chapter 5) built upon study 2 and incorporated changes to the research design (e.g., use of multidimensional scaling, logistic regression) to map the factors and indicators to a larger sample of ISIS foreign fighters and sympathisers. The fourth study (chapter 6) was designed to replicate the research methodology of study 3 using a different sample (i.e., Kurdish fighters) to compare against ISIS fighters. Finally, the fifth study (chapter 7) detailed a survey conducted with members of the community (NTU undergraduates and MTurk workers) on their perceptions towards the indicators and their general understanding of online radicalisation and terrorism. The practical, methodological, and theoretical implications of this research as well as suggestions for future studies are discussed in chapter 8.
... A recent study reported by Kalpakis et al., 43 was based on analyzing speci¯c traits of terrorism-related content posted on Twitter. They aimed at distinguishing terrorism-related twitter accounts from others. ...
Full-text available
With the ever-growing number of online social media platforms, the world has shrunk even further with regards to communication and knowledge-sharing perspective. However, communication, at times, can be deterrent when misused using such widespread social media tools. The acts of terrorism become seemingly convenient as the barrier of communication is nullified. This propagation of hateful content becomes much more easier and even recruiting anti-socials gets easier. Oddly, these social media platforms are the ones that prove essential during such crisis situations. This paper reviews most of the works reported by various authors in the last 10 years on the use of social media during a time of terrorist attack to addressing how to use social media for public communication with the emergency organization and military or police during terrorist attack, how to perform post-attack social media analytic and how to detect acts of terrorism, unrest, and hatred using social media analytic. With this objective, the authors also hope to inspire other researchers to work in this direction and use this review as a guide for instigating future research to counter-attack terrorism as it is the need of the hour for our country in the wake of recent Uri and Pulwama attack.
Conference Paper
Full-text available
Monitoring and analysis of web forums is becoming important for intelligence analysts around the globe since terrorists and extremists are using forums for spreading propaganda and communicating with each other. Various tools for analyzing the content of forum postings and identifying aliases that need further inspection by analysts have been proposed throughout literature, but a problem related to this is that individuals can make use of several aliases. In this paper we propose a number of matching techniques for detecting forum users who make use of multiple aliases. By combining different techniques such as time profiling and stylometric analysis of messages the accuracy of recognizing users with multiple aliases increases, as shown in experiments conducted on the ICWSM dataset
Conference Paper
Full-text available
The paper describes a technical demonstration of an open-source framework for monitoring, analysis and search over multiple social media platforms. The framework is intended to be a valuable tool for media intelligence professionals, as well as a framework and testbed for scientists and developers with interest in social media research.
Full-text available
The large-scale availability of user-generated content in social media platforms has recently opened up new possibilities for studying and understanding the geospatial aspects of real-world phenomena and events. Yet, the large majority of user-generated content lacks proper geographic information (in the form of latitude and longitude coordinates). As a result, the problem of multimedia geotagging, i.e., extracting location information from user-generated text items when this is not explicitly available, has attracted increasing research interest. Here, we present a highly accurate geotagging approach for estimating the locations alluded by text annotations based on refined language models that are learned from massive corpora of social media annotations. We further explore the impact of different feature selection and weighting techniques on the performance of the approach. In terms of evaluation, we employ a large benchmark collection from the MediaEval Placing Task over several years. We demonstrate the consistently superior geotagging accuracy and low median distance error of the proposed approach using various data sets and comparing it against a number of state-of-the-art systems.
Conference Paper
Full-text available
Social media are widely used among terrorists to communicate and disseminate their activities. User-to-user interaction (e.g. mentions, follows) leads to the formation of complex networks, with topology that reveals key-players and key-communities in the terrorism domain. Both the administrators of social media platforms and Law Enforcement Agencies seek to identify not only single users but groups of terrorism-related users so that they can reduce the impact of their information exchange efforts. To this end, we propose a novel framework that combines community detection with key-player identification to retrieve communities of terrorism-related social media users. Experiments show that most of the members of each retrieved key-community are already suspended by Twitter, violating its terms, and are hence associated with terrorism-oriented content with high probability.
Full-text available
Growing use of the Internet as a major means of communication has led to the formation of cyber-communities, which have become increasingly appealing to terrorist groups due to the unregulated nature of Internet communication. Online communities enable violent extremists to increase recruitment by allowing them to build personal relationships with a worldwide audience capable of accessing uncensored content. This article presents methods for identifying the recruitment activities of violent groups within extremist social media websites. Specifically, these methods apply known techniques within supervised learning and natural language processing to the untested task of automatically identifying forum posts intended to recruit new violent extremist members. We used data from the western jihadist website Ansar AlJihad Network, which was compiled by the University of Arizona’s Dark Web Project. Multiple judges manually annotated a sample of these data, marking 192 randomly sampled posts as recruiting (Yes) or non-recruiting (No). We observed significant agreement between the judges’ labels; Cohen’s κ=(0.5,0.9) at p=0.01. We tested the feasibility of using naive Bayes models, logistic regression, classification trees, boosting, and support vector machines (SVM) to classify the forum posts. Evaluation with receiver operating characteristic (ROC) curves shows that our SVM classifier achieves an 89% area under the curve (AUC), a significant improvement over the 63% AUC performance achieved by our simplest naive Bayes model (Tukey’s test at p=0.05). To our knowledge, this is the first result reported on this task, and our analysis indicates that automatic detection of online terrorist recruitment is a feasible task. We also identify a number of important areas of future work including classifying non-English posts and measuring how recruitment posts and current events change membership numbers over time.
Social media have played an essential role in the jihadists’ operational strategy in Syria and Iraq, and beyond. Twitter in particular has been used to drive communications over other social media platforms. Twitter streams from the insurgency may give the illusion of authenticity, as a spontaneous activity of a generation accustomed to using their cell phones for self-publication, but to what extent is access and content controlled? Over a period of three months, from January through March 2014, information was collected from the Twitter accounts of 59 Western-origin fighters known to be in Syria. Using a snowball method, the 59 starter accounts were used to collect data about the most popular accounts in the network-at-large. Social network analysis on the data collated about Twitter users in the Western Syria-based fighters points to the controlling role played by feeder accounts belonging to terrorist organizations in the insurgency zone, and by Europe-based organizational accounts associated with the banned British organization, Al Muhajiroun, and in particular the London-based preacher, Anjem Choudary.
Tweeting propaganda, radicalization and recruitment: Islamic state supporters multi-sided twitter networks
  • A T Chatfield
  • C G Reddick
  • U Brajawidagda
Chatfield, A. T., Reddick, C. G., & Brajawidagda, U. (2015). Tweeting propaganda, radicalization and recruitment: Islamic state supporters multi-sided twitter networks. In Proceedings of the 16th Annual International Conference on Digital Government Research (pp. 239-249).
Twitter suspends 377,000 accounts for pro-terrorism content
  • S Larson
Larson S. (2017). Twitter suspends 377,000 accounts for pro-terrorism content.http:// Accessed 15 Sept 2017.
Tweeting the Jihad: Social media networks of western foreign fighters in Syria and Iraq
  • J Klausen
Klausen, J. (2015). Tweeting the Jihad: Social media networks of western foreign fighters in Syria and Iraq. Studies in Conflict & Terrorism, 38(1), 1-22.