Abstract
The development of social media technologies has had a significant impact on individuals, organizations and societies. However, social media has not only affected people and communities, but also terrorist organizations have started to use social media platforms effectively. The Islamic State of Iraq and Syria (ISIS) is one such group that actively utilizes social media. Social media has played a critical role in formulating and disseminating ISIS’s strategies. Twitter is one of the most effective social media platforms used by ISIS, and it actively uses Twitter in predominantly Muslim countries such as Turkey. In this article, I use a dictionary-based method to compare and analyze 29,419 tweets from English-speaking supporters of ISIS and 40,526 tweets from Turkish supporters of ISIS between 2015 and 2016.
1 Introduction1
The world has undergone a profound transformation with the advent of mass communication, artificial intelligence, the internet and social media technologies that emerged in the 2000s (Subramanian 2018). Of these, the internet and social media technologies have significantly impacted people’s lives, becoming an integral part of their daily routines (Lazer et al. 2009). This is mainly due to the speed and ease with which people can access and share information through these technologies. Furthermore, social media and internet technologies enable individuals to express their opinions and feelings on various issues to thousands of people across different countries in a short time (Lazer et al. 2009; Watts 2007; Adamic and Glance 2005).
However, not just individuals have taken advantage of the internet and social media, but also terrorist organizations. Social media and the internet have become essential tools for terrorist groups to propagate their ideologies, propaganda and violent actions (Gaikwad et al. 2021; Lieberman 2017). Scholars such as Clark (2016) and Barceló and Labzina (2020) suggest that the internet and social media allow terrorist organizations to disseminate their violent images and videos without censorship, falsehoods or exaggeration. According to Marin (2016) and Kaakinen et al. (2021), today’s terrorist organizations can influence not only the immediate victims of terrorist attacks but also those who are exposed to the groups’ posts on the internet and social media by using the internet and social media tools (e.g., Redmond 2017; Hoffman 2017).
The active use of the internet and social media platforms, such as Facebook and Twitter, by terrorist organizations in their strategies has drawn the attention of the academic community. Consequently, many academic studies have examined the methods used by terrorist organizations on social media platforms, such as Facebook (e.g., Waters and Postings 2018; Ayad 2020) and Twitter (e.g., Lee and Colautti 2022; Ashcroft et al. 2015). The Islamic State of Iraq and Syria’s (ISIS) active use of social media platforms and internet technologies in its strategies has resulted in a focus on ISIS in the academic literature on the relationship between terrorist organizations and social media or the internet (e.g., Steadman 2020; Vujanovic 2016). Since Twitter is one of the most actively used social media platforms by ISIS, numerous academic studies have approached ISIS’s Twitter activities from various perspectives (e.g., Tulga 2022b; Lee and Colautti 2022; Leite et al. 2019; Winter 2017).
Accordingly, some studies focus on detecting ISIS sympathizers on Twitter (e.g., Ashcroft et al. 2015; Bodine-Baron et al. 2016; Berger and Morgan 2015; Conway et al. 2019; Conway 2017). Other studies try to find the main actors among ISIS sympathizers on Twitter and focus on the characteristics of their tweets (Fernandez, Asif and Alani 2018; Klausen 2015; Chatfield, Reddick and Brajawidagda 2015). The remaining studies focus on the number of Twitter followers and tweets of ISIS sympathizers (Abd-Elaal, Badr and Mahdi 2020; Berger and Perez 2016; Berger 2015).
Despite numerous studies on the use of Twitter by ISIS, most of them have focused on the content of tweets by ISIS sympathizers (e.g., Dillon, Neo and Freilich 2020; Ekwenchi, Ononiwu and Efetobor 2016), the topics they discuss on Twitter (e.g., Al-Rawi and Groshek 2018) and the network of ISIS supporters on Twitter (e.g., Al-Khateeb and Agarwal 2015; Klausen 2015). Some studies in this area suggest that the contents of tweets by ISIS sympathizers are related to religious topics (e.g., Neo et al. 2017; Ali 2014), while others argue that they are connected to violence or war (e.g., Dillon, Neo and Freilich 2020). Although these studies primarily focus on the tweets of foreign sympathizers of ISIS in English (e.g., Guthrie 2018), some studies analyze the tweets of ISIS sympathizers in Arabic (e.g., Ceron, Curini and Iacus 2019; Al-Rawi and Groshek 2018). However, few studies have compared the content of tweets by ISIS sympathizers in a predominantly Muslim country such as Turkey with those of ISIS supporters in English.
This study aims to answer a fundamental research question: “Are the general characteristics and content of tweets by Turkish and English-speaking sympathizers of ISIS different?” To address this question, I analyze and compare tweets from Turkish and English-speaking supporters of ISIS during 2015–2016. This study uses a dictionary-based text analysis approach to analyze 29,419 English and 40,526 Turkish tweets from Turkish and English-speaking ISIS sympathizers.
The next section of this article presents the findings of various studies in the literature. Firstly, the article gives descriptive information on Turkish and English-speaking ISIS supporters, such as the number of tweets and users who tweet the most. Secondly, it analyzes the supporters’ tweets with the dictionary-based text analysis method. The processes of these analyses are explained in the research design section. The results of the analysis of tweets’ contents by Turkish and English-speaking ISIS sympathizers are presented in detail in the results section. The discussion section explains the key findings and highlights similarities and differences with previous studies in the literature. Finally, the article concludes with a summary of the main points.
2 Literature Review
Contemporary terrorist organizations understand the power of the internet and social media channels (Näsi et al. 2020; Oksanen et al. 2020). For this reason, numerous scholars argue that today’s terrorist organizations actively use social media and the internet in their strategies (e.g., Pantucci 2011; Finseraas and Listhaug 2013; Windsor 2020). These scholars argue that terrorist organizations aim to expand their activities using the internet and social media strategies (Hoffman 2017; Seib and Janbek 2010; Tulga 2022a). For example, Weinburg and Eubank (2007) think that modern forms of communication (i.e., the internet) promote the spread of terrorism from one place to another because social media and the internet have obliterated the barriers of time and space.
ISIS is a prominent example of a terrorist organization that actively uses modern forms of communication, including the internet and social media platforms. By using various internet resources, including social media, video platforms, and professionally created pictures, magazines and propaganda movies, ISIS has successfully transformed itself into a highly skilled propaganda machine (Özeren et al. 2018). With its social media and internet strategies, ISIS aims to win the hearts and minds of young generations of Muslims worldwide while creating an atmosphere of fear in groups it considers the “enemy” (Chatfield, Reddick and Brajawidagda 2015).
For this reason, ISIS’s use of social media has attracted the attention of many scholars (e.g., Hoffman 2017; Shrira 2015). Many such studies focus on the content of social media posts by ISIS sympathizers (e.g., Msall and Lary 2022; Gomes et al. 2017). While some of these studies analyze the social media posts of sympathizers with simple text analysis methods, such as the most used words, others examine the social media posts of sympathizers with more complex text analysis methods, such as topic modeling or dictionary-based approaches.
It is possible to divide the findings revealed by all these methods into two main views. Some studies argue that the content of social media posts by ISIS sympathizers has religious or justification content (e.g., El-Badawy, Comerford and Welby 2015). In contrast, others say that sympathizers post about violence, terrorism or war (e.g., Al-Rawi and Groshek 2018).
In the first category, scholars argue that the social media posts of ISIS sympathizers have religious or justification content (Rehman et al. 2021; Vergani and Bliuc 2015). Gomes and colleagues (2017) analyzed the tweets of ISIS sympathizers using the most used word method. As a result of their analysis, the authors stated that ISIS sympathizers mainly used religious words and mostly shared religious content. The authors found that the words “God” and “Allah” were the most frequently used in tweets (Gomes et al. 2017). Similarly, Rehman and colleagues (2021) found that religion-related words such as “Allah,” “Muslim,” “Prophet” and “Islam” were the most frequently used words in tweets of ISIS sympathizers.
Meanwhile, some studies in the first category have employed relatively more sophisticated text analysis methods to examine the social media posts of ISIS sympathizers, going beyond simple text analysis methods. For instance, Vergani and Bliuc (2015) used automated text analysis methods to examine ISIS’s online propaganda. The authors discovered that ISIS employs emotional sentences and religious quotes on social media platforms. Similarly, Bodine-Baron et al. (2016) found that the tweets of ISIS supporters conveyed positive themes such as religion and a sense of belonging.
Scholars in the second category argue that ISIS sympathizers’ social media posts contain content related to violence, war or terrorism (Ashcroft et al. 2015; Zahrah, Nurse and Goldsmith 2020). Some such studies have examined the social media posts of ISIS supporters using the most frequently used words method. Ashcroft and colleagues (2015) found that ISIS sympathizers frequently use terms associated with violence, hostility and war, such as “kill,” “state” and “infidel,” in their tweets and that supporters generally tweet negative content related to violence. Similarly, Lorenzo-Dus, Kinzel and Walker (2018) found that ISIS frequently uses words such as “disbelievers” and “apostates” to refer to non-Muslim groups on online platforms and urges sympathizers to engage in violence against non-Muslim groups.
Other studies in the second category have employed diverse text analysis methods to examine the social media posts of ISIS sympathizers, supporting the contention that these posts are related to violence. Using the topic modeling method, Zahrah, Nurse and Goldsmith (2020) analyzed the tweets of ISIS sympathizers and found that the most common topic discussed was reporting on the latest attacks by and against ISIS. Similarly, Msall and Lary (2022) conducted a content analysis of ISIS posts on online platforms and identified themes of injustice, “connecting actions to experience” and “call to action.” Finally, Agarwal and colleagues (2017) employed the k-nearest neighbors (k-NN) and Support Vector Machine (SVM) algorithms to analyze tweets from ISIS sympathizers. They found that these individuals tweeted about war with negative and offensive content.
Similar studies have also focused on the content of tweets by Turkish sympathizers of ISIS. However, most of these studies were analyzed using simple text analysis methods, such as the most frequently used words or qualitative methods. One study by Özeren and colleagues (2018) found that based on the most commonly used words, the most discussed topics by Turkish sympathizers of ISIS were Jihad, emigrating (Hijra), PKK (Partiya Karkeren Kurdistane) and YPG (Yekineyen Parastine Gel). Similarly, Yalçınkaya (2016) found that the most frequently used words by Turkish sympathizers of ISIS were bomb, YPG and PKK. Using a qualitative method for content analysis, the author discovered that the topic that supporters discussed the most was related to the PKK (Yalçınkaya 2016). Finally, using a qualitative method, Narin and Ayaz (2016) analyzed 509 tweets from ten Turkish sympathizers of ISIS. The authors argued that sympathizers consider the PKK and the Republic of Turkey as enemies and described them with words such as “atheist” and “infidel” (Narin and Ayaz 2016).
Despite the valuable research in the literature, there is a limited number of studies examining the differences and similarities between the content of tweets by foreign supporters of ISIS and those of ISIS sympathizers in predominantly Muslim countries such as Turkey. Moreover, studies focusing on Turkish sympathizers of ISIS tend to use simple text analysis or qualitative methods, which may not provide detailed information on their tweets. Thus, this study aims to compare the content of tweets by English-speaking supporters of ISIS in English with those of Turkish supporters of ISIS between 2015 and 2016.
3 Research Design
In this study, I used Twitter data to compare the contents of tweets by English-speaking sympathizers of ISIS in English with those of Turkish sympathizers of ISIS. Specifically, I aimed to highlight the contextual differences or similarities present in the tweets of these two groups. The most important reasons for choosing Twitter in this study are the free access to Twitter data, unlike social media platforms such as Facebook and Instagram, and the active use of Twitter by Turkish and English-speaking sympathizers of ISIS.
Twitter is a popular social media platform that provides users an online space to discuss and share opinions on various topics (Sriram et al. 2010; Zahrah, Nurse and Goldsmith 2020). As of 2022, more than 396 million people worldwide used Twitter, and over 500 million tweets were posted daily (Kemp 2022). Twitter is widely used by people globally to share information or news about their daily lives and express their feelings (Ficamos and Liu 2016). However, terrorist organizations have also exploited Twitter to disseminate their violent propaganda videos and propagate their ideology (Bodine-Baron et al. 2016). Therefore, examining the large pool of Twitter activities of people and terrorist organizations as researchers is crucial to contributing to the literature on terrorism.
Twitter is a powerful social media platform in Turkey, just as it is globally. According to recent data, 34 percent of the Turkish population uses Twitter to obtain news and information (Newman 2022; Kemp 2022). Moreover, ISIS and its supporters use Twitter effectively in Turkey for several purposes, such as recruitment (Özeren et al. 2018) and creating fear among people (Yüksel and Yıldırım 2016).
However, there is limited research in the literature that compares the content of tweets of foreign sympathizers of ISIS with those of ISIS sympathizers in predominantly Muslim countries such as Turkey (e.g., Colombo and Curini 2022; Bodine-Baron et al. 2016). Therefore, the differences or similarities between the Twitter content of ISIS sympathizers in predominantly Muslim countries and the Twitter content of foreign sympathizers of ISIS remain unclear.
Thus, it is crucial to compare the tweets of Turkish ISIS sympathizers with those of English-speaking ISIS supporters. The selected tweets were posted between January 2015 and December 2016. The study analyzed 40,526 Turkish tweets and 29,419 English tweets from ISIS supporters.
However, the first and foremost step in this investigation was to gather the tweets of ISIS sympathizers, both Turkish and English-speaking. To obtain these tweets, the Twitter API was utilized.2 This API provides users with access to data related to tweets, retweets and favorites. The R programming language also allows for using this API and its analysis. One package that enables the collection of tweets is the “twitteR” package, which I used in this study (Gentry 2016). Tweet collection using the Twitter API began in January 2015.3 Tweet collection continued until January 2017.
Later, I found tweets in English and Turkish on hashtags such as “#ISIS,” “#Raqqa (Raqqa),” “#Dawlah (state),” “#Haseke,” “#Holkampı (Holcamp),” “#haçlıkoalsiyonunahayır (notocrusadercoalition),” “#fridayinaleppo,” “#islamicstate,” “#caliphatefollow,” “#totheislamicstate” and “#mohammedyoussefthelion.”4 However, these hashtags were used not only by ISIS sympathizers but also by anti-ISIS users. Conway and colleagues (2019) stated that ISIS sympathizers could be identified by their profile photos, avatar photos, user handles and screen names. For this reason, I separated the ISIS sympathizers from other users with an algorithm. I included profile photos, avatar photos, user handles and screen names in the algorithm. I randomly selected the tweets obtained by the algorithm and double-checked whether the tweets were in support of ISIS.5 In the following step, I divided the English and Turkish tweets into two separate data sets for analysis.
The collected data from Twitter comprises 90 variables, such as user name, account number, location, number of followers, the tool used to send the tweets (e.g., mobile Twitter or website), retweets and original tweets. However, since not all variables are used in the study, I focused on nine variables, including user identification, user screen name, time of tweets, original tweet, location of users, number of retweets, number of favorites, tool or source used to send the tweets and number of followers. The data set contains 29,419 observations in English tweets and 40,526 in Turkish tweets. The unit of analysis for this Twitter examination is tweets. However, to provide descriptive information in the study, the users who tweeted the most among Turkish and English-speaking supporters of ISIS were also analyzed.
Once the tweets of Turkish and English-speaking ISIS sympathizers were collected, the analysis began. I examined the Turkish and English-speaking sympathizers of ISIS in four main stages. First, the number of tweets of ISIS sympathizers was analyzed. In addition to the number of tweets, I determined the years and months in which the Turkish and English-speaking sympathizers of ISIS tweeted the most.
In the second step, I identified the most commonly used words by ISIS sympathizers, both Turkish and English-speaking. This information can provide insights into the topics that sympathizers often discuss. To achieve this, it is essential to clean the tweets by removing stop words, mentions, punctuation, numbers, hashtags, URL s and emojis. In addition, duplicate tweets were removed from the data set to minimize the “noise” levels. At the end of the cleaning step, the sentences were split into tokens—this step is called tokenization (Harbi and Kamsin 2021). Usually, tokens are words because these are the most common semantically meaningful components of texts (Welbers, Atteveldt and Benoit 2017). When processing data, “letter” and “letters” could be treated as distinct features. This stage effectively allows the removal of plurals and inflections (Ficamos and Liu 2016), which is called stemming (Omer 2015). As a result, the unigrams (one word) feature can be lemmatized to help in the process of reduction (Mitts, Phillips and Walter 2022). After cleaning, my data set containing tweets from Turkish and English-speaking ISIS sympathizers became ideal for finding the most frequently used words.
After identifying the most commonly used words, a dictionary-based analysis was conducted. To perform this analysis, two dictionaries were prepared for “terrorism/violence” and “religion/justification.” The main reason for separating the analysis into two dictionaries is the findings in the literature in this direction (e.g., Badawy and Ferrara 2018). For instance, Badawy and Ferrara (2018) discovered that two subjects dominate ISIS discussions and that ISIS members mainly discuss topics connected to violence and Islamic theology. The dictionary-based analysis involves searching records for specific words and generating a prevalence score for each record for each term (Bonikowski and Gidron 2015; Nouh, Nurse and Goldsmith 2019; Badawy and Ferrara 2018).
A manual list was compiled to create the dictionaries, including relevant terms related to “terrorism/violence” and “religion/justification.” This list was generated using resources such as the Cambridge and Merriam-Webster dictionaries and the WordNet software program, which provides synonyms and antonyms for selected words. Additionally, as suggested by Bonikowski and Gidron (2015), the extensive literature on recruitment, propaganda, religion, terrorism and violence was reviewed to identify potentially relevant terms.
The “terrorism/violence” dictionary contains 401 words, including terms such as crusader, destroy, disobedience, disbeliever, traitor and hell. The “religion/justification” dictionary also comprises 401 words, including words such as “amir,” “awliya,” “Baqarah” and “believer.” The dictionaries’ creation enabled a comprehensive dictionary-based data analysis, essential for understanding the topics that ISIS sympathizers discuss on Twitter.
Both dictionaries have been translated into Turkish, and exact translations of the words are used in both dictionaries. These two dictionaries are clean of terms that generated an overwhelming number of false positives. All words in the dictionaries are unigrams. To ensure the robustness of the dictionary-based analysis, the dictionaries were randomly divided into two, and the investigation was repeated. This stage is known as the “split-half test” in the literature. Similar results were obtained at the end of the robustness test, indicating the analysis is reliable.
Colombo and Curini (2022) observed that the tweet content of ISIS supporters from October 2014 to September 2017 varied depending on the military successes or losses of ISIS. Accordingly, in the fourth and final stage of my study, I conducted separate analyses for 2015 and 2016 to examine if the content of tweets by Turkish and English-speaking ISIS sympathizers had changed over the years.
After conducting these four steps, I determined the number of tweets and the most active users who tweeted them. Then, I identified the words most commonly used by Turkish and English-speaking supporters of ISIS. Lastly, I analyzed the content of the tweets posted by these sympathizers and investigated whether there had been any changes in their tweets over the years.
4 Results
First, I give descriptive information about Turkish and English-speaking ISIS sympathizers on Twitter. In this context, I explain how many users tweet, how many tweets by users and who tweets the most. The number of ISIS sympathizers tweeting in English is 112 in the data set: 61.61 percent of these users have zero to 99 tweets; 33.03 percent have between 100 and 499 tweets; and 5.36 percent have between 500 and 1,571 tweets (Figure 1).
Upon analyzing the Turkish tweets of ISIS sympathizers, it was found that 350 users tweeted. The analysis revealed that 76.28 percent of these users had zero to 99 tweets; 18.57 percent had 100 to 499 tweets; and 5.14 percent had 500 to 4,380 tweets (Figure 2).
The examination of the English tweets regarding ISIS reveals that the user who tweeted the most has 1,571 tweets, and the user with the most tweets is “@Uncle_SamCoco.” It has been found that the five users with the most tweets have between 1,000 and 1,571 tweets (Figure 3).
When the Turkish tweets of ISIS sympathizers are examined, I find that the user who tweeted the most has 4,380 tweets, and the user with the most tweets among Turkish tweets is “@KartaL_Efe42.” The five users with the most tweets have between 1,400 and 4,380 tweets (Figure 4). Based on these results, it has been shown that the Turkish sympathizers of ISIS who tweet the most use Twitter more actively.
When we look at the years when ISIS sympathizers tweeted in English and Turkish, it is clear that 2016 saw the most tweets in English, while 2015 saw the most tweets in Turkish. The highest number of tweets in English were posted in April 2016 (Figure 5), while the highest number of Turkish tweets were posted in January 2015 (Figure 6).
In the last step of the study, which examined the content of tweets by ISIS sympathizers in Turkish and English, the most frequently used words in Turkish and English tweets by ISIS sympathizers were found. As Ashcroft et al. (2015) and Gomes et al. (2017) have shown, the most commonly used words in tweets can provide information about the content of the tweets. Therefore, in this phase of the research I first determined the most frequently used words in the Turkish and English tweets.
Figure 1
Number of tweets in English (as a percentage)
Citation: Innovation in the Social Sciences 1, 2 (2023) ; 10.1163/27730611-bja10013
Source: Processed by the authorFigure 2
Number of tweets in Turkish (as a percentage)
Citation: Innovation in the Social Sciences 1, 2 (2023) ; 10.1163/27730611-bja10013
Source: Processed by the authorFigure 3
Most tweeting English-speaking supporters
Citation: Innovation in the Social Sciences 1, 2 (2023) ; 10.1163/27730611-bja10013
Source: Processed by the authorI found that the five most frequently used words in the English tweets are “ISIS,” “Syria,” “kill,” “IS,” and “USA (Figure 7).” In addition, I found that only “Allah” is used among the 25 most commonly used words related to religion. An examination of the most frequently used words shows that ISIS sympathizers use words related to ISIS, violence and war rather than religious words in their English tweets.
Figure 4
Most tweeting Turkish supporters
Citation: Innovation in the Social Sciences 1, 2 (2023) ; 10.1163/27730611-bja10013
Source: Processed by the authorFigure 5
Number of tweets by months/years (English)
Citation: Innovation in the Social Sciences 1, 2 (2023) ; 10.1163/27730611-bja10013
Source: Processed by the authorFigure 6
Number of tweets by months/years (Turkish)
Citation: Innovation in the Social Sciences 1, 2 (2023) ; 10.1163/27730611-bja10013
Source: Processed by the authorIn contrast, when we look at the words most frequently used by ISIS sympathizers in their Turkish tweets, the most commonly used word is “Allah (god) (Figure 8).” Words such as “Ahrar al-Sham,” “Islam,” “Asad,” “Muslim” and “PKK” are among the six most frequently used words. Unlike the most commonly used words in English tweets, words related to religion are frequently used in Turkish tweets, but in addition to religious words, words such as “Assad” and “PKK” related to war, and “Islamic Front” and “army of conquest” associated with ISIS, appear to be used frequently.
Figure 7
Most frequently used words in English tweets
Citation: Innovation in the Social Sciences 1, 2 (2023) ; 10.1163/27730611-bja10013
Source: Processed by the authorFigure 8
Most frequently used words in Turkish
Citation: Innovation in the Social Sciences 1, 2 (2023) ; 10.1163/27730611-bja10013
Source: Processed by the authorThus, words such as “Allah (God),” “Assad,” “USA,” “News” and “PKK/YPG” are used frequently in both Turkish and English tweets.
As some studies in the literature show (e.g., Badawy and Ferrara 2018; Al-Khateeb and Agarwal 2015), ISIS members mainly discuss issues related to two main topics—violence and Islamic theology—and therefore a dictionary-based method focusing on these two main topics is also used in this study. My dictionary-based analysis has shown that 61.17 percent of English tweets by ISIS sympathizers contain violent or terrorism content, while 38.82 percent have religious or justification content (Figure 9).
Figure 9
Dictionary-based results of English tweets
Citation: Innovation in the Social Sciences 1, 2 (2023) ; 10.1163/27730611-bja10013
Source: Processed by the authorFigure 10
Dictionary-based results of Turkish tweets
Citation: Innovation in the Social Sciences 1, 2 (2023) ; 10.1163/27730611-bja10013
Source: Processed by the authorIn contrast, the analysis has shown that 55.17 percent of the Turkish tweets of ISIS sympathizers have religious or justification content, while 44.82 percent have violence or terrorism content (Figure 10).
Both results are consistent with the results of the most frequently used words by ISIS sympathizers in English and Turkish tweets. However, the analysis is taken a step further and focuses on the change in the content of tweets over the years, as Colombo and Curini (2022) have suggested.
As a result of the analysis by year, I find that in 2015, 50.84 percent of the English tweets of ISIS sympathizers had violence or terrorism content and 49.15 percent had religious or justification content. In the same year, 43.12 percent of Turkish tweets contained violence or terrorism content and 56.88 percent had religious or justification content (Figure 11).
However, I find that the content of violence or terrorism in tweets from English-speaking ISIS sympathizers increased to 61.72 percent and religious or justification content decreased to 38.28 percent in 2016. Similarly, I find that more than 50.89 percent of Turkish tweets had religious or justification content, as in 2015, but the content of terrorism or violence increased to 49.11 percent in 2016. Thus, the tweets of Turkish and English-speaking ISIS sympathizers changed over time in terms of content (Figure 11).
Figure 11
Comparison of Turkish and English tweets’ content
Citation: Innovation in the Social Sciences 1, 2 (2023) ; 10.1163/27730611-bja10013
Source: Processed by the author5 Discussion
Many significant findings have been reached as a result of this study. While some of these findings are consistent with the literature, others differ from it. The first finding is that while there are 112 users in the data set containing the English-speaking tweets of ISIS sympathizers, there are 350 users in the data set containing the Turkish tweets of ISIS sympathizers. The five users who tweeted the most in English have between 1,000 and 1,571 tweets, while the five users who tweeted the most in Turkish had between 1,400 and 4,380 tweets. Thus, the Turkish sympathizers of ISIS who tweeted the most were more active on Twitter than the English-speaking sympathizers of ISIS who tweeted the most.
Upon analyzing the number of tweets per user in both data sets, it was found that most users in the English data set had tweeted fewer than 100 times. Similarly, 76.28 percent of users in the Turkish data set had tweeted fewer than 100 times. These results are consistent with some previous studies in the literature, such as Gomes et al. (2017), who also found that most ISIS sympathizers tweeted fewer than 100 times. However, the findings of my study differ from those of Berger and Perez (2016), who reported that, on average, ISIS sympathizers tweeted at least 200 times. In contrast, my study found that 33.03 percent of English-speaking ISIS sympathizers and 18.57 percent of Turkish ISIS sympathizers tweeted an average of 200 tweets.
Upon examining the years during which Turkish and English-speaking ISIS sympathizers tweeted the most, it was discovered that Turkish sympathizers were most active in 2015, with the highest activity observed in January 2015. Conversely, English-speaking ISIS sympathizers were most active in 2016, with the most increased activity recorded in April 2016.
In my study’s final step, which focuses on the tweet content of Turkish and English-speaking sympathizers of ISIS, I reach essential findings. To provide partial information about the content, I have examined the most frequently used words in tweets. I found that the most commonly used words in English tweets by English-speaking sympathizers of ISIS are “ISIS,” “Syria,” “Kill” and “USA.” “Allah” is the only word related to religion in the 25 most frequently used terms. These findings are consistent with previous research. Ashcroft et al. (2015) reported that the most commonly used words in ISIS tweets included “Syria,” “kill” and “Allah.” Similarly, Gomes et al. (2017) found that ISIS sympathizers frequently used the word “Allah” in connection with religion.
In contrast to the frequently used words in English tweets by English-speaking sympathizers of ISIS, I found that Turkish sympathizers of ISIS often used terms associated with religion, such as “Allah,” “Muslim” and “Islam,” in their tweets. Additionally, I discovered that frequently used words related to war and “out-groups,” such as “PKK,” “Asad” and “Infidel,” are used by them. This finding is somewhat different from some previous studies. Yalçınkaya (2016) discovered that Turkish sympathizers of ISIS commonly used words such as “bomb,” “PKK” and “YPG,” and religious terms were not frequently used. Similarly, Narin and Ayaz (2016) analyzed 509 tweets of ISIS sympathizers and found that use of the word “USA” was limited in the tweets. In contrast, my analysis results are consistent with those of Özeren, Çubukçu and Cash (2021), who found that Turkish sympathizers of ISIS frequently use words related to religion in their tweets.
However, relying solely on the most frequently used words may not provide in-depth information about the content of tweets. Therefore, to gain a deeper understanding of the content of the tweets, I used a dictionary-based analysis method to examine the tweets of Turkish and English-speaking sympathizers. The analysis revealed that 61.17 percent of the English tweets by English-speaking ISIS sympathizers contained violent or terrorist content, while 38.28 percent contained religious or justificatory content. These findings are in line with previous studies. For instance, Al-Khateeb and Agarwal (2015) reported that ISIS social media propaganda contained offensive, war and hate speech. Additionally, Zahrah, Nurse and Goldsmith (2020) found that ISIS sympathizers frequently shared news about the group’s war victories and expressed negative views about perceived enemies on social media platforms. However, my study’s results contradict the findings of some studies in the literature. For example, El-Badawy, Comerford and Welby (2015) found that ISIS sympathizers’ social media posts mainly focused on religious content, including quotes from the Quran and hadiths, to legitimize the group.
Upon analyzing the tweets of Turkish sympathizers of ISIS, it was found that 55.17 percent contained religious or justification content, while 44.82 percent contained violent or terrorist content. These results are consistent with the findings of studies focusing on ISIS sympathizers in Turkey. For example, Özeren et al. (2018) found that ISIS uses Twitter for recruitment purposes in Turkey and that sympathizers actively tweet for this purpose. Similarly, Emir Türkoğlu conducted a qualitative study on the Konstantiniyye, an ISIS magazine in Turkey, and found that while the magazine predominantly uses religious language, it also contains hidden messages of violence and threats underneath the religious discourse (Türkoğlu 2017).
ISIS has targeted Turkey for recruitment purposes due to its predominantly Muslim population, as noted by Özeren et al. (2018). Through Twitter, ISIS calls on Muslims in Turkey to migrate to the lands under its control. Meanwhile, English-speaking sympathizers of ISIS on Twitter primarily aim to create an atmosphere of fear among people in countries in Europe and North America, which ISIS considers its enemies, according to Hoffman (2017). To achieve this goal, they primarily use words related to violence or terrorism in their English tweets.
In the last stage of the study, I focus on whether the content of tweets by English-speaking and Turkish sympathizers of ISIS changed between 2015 and 2016. Colombo and Curini (2022) analyzed the tweets of ISIS sympathizers in three periods between 2014 and 2017 and named these three periods as peak, decline and defeat. The year 2015, covered by my study, is included in the period that the authors call the decline period. The year 2016 is the period the authors call the defeat period. As a result of my research, I discovered that in 2015, 50.84 percent of English-speaking ISIS sympathizers’ tweets in English contained violence or terrorism content, while 49.15 percent contained religious or justification content. However, with the defeat period that started in 2016, sympathizers increased the content related to violence or terrorism.
Similarly, while 56.88 percent of Turkish tweets had religious or justification content in 2015, this rate decreased to 50.89 percent in 2016, and tweets containing violence or terrorism content increased to 49.11 percent in 2016. Although ISIS viewed Turkey as a recruitment base, the main reason for the rise in violence or terrorism content in the tweets was Turkey’s increasing involvement in the war against ISIS in several fields by 2016. As part of the fight against ISIS, Turkey opened the Incirlik Airbase for the use of crewed and uncrewed aerial vehicles by the anti-ISIS coalition towards the end of 2015. Additionally, Turkey officially became a member of the anti-ISIS coalition on 24 August 2015, and the first joint air campaign was carried out on 28 August. Furthermore, Turkey initiated operations in northern Syria on 26 August 2016 to thwart ISIS attacks, stop migration from Syria and eliminate ISIS elements in the region. The significant role played by Turkey against ISIS was also reflected in the tweets of Turkish sympathizers, with the violence or terrorism discourse in the content of the tweets increasing to 49.11 percent.
Despite the study’s essential and noteworthy findings, there are also some critical limitations. The most important of these limitations is the accounts of ISIS sympathizers that the Twitter company deactivated. Due to the deactivation of the accounts, the tweets of these accounts and general information about the users could not be accessed. Therefore, these users’ tweets could not be included in the analysis.
6 Conclusion
In recent years, social media platforms such as Twitter have experienced dramatic growth and have become an integral part of people’s daily lives. Twitter and other social media platforms offer an important opportunity for millions of users to share information and express their views on various issues. Moreover, social media platforms provide a platform for millions of people from different countries and regions to quickly share their feelings and reactions to different events and issues. However, social media platforms are used not only by ordinary people but also by terrorist groups. Many terrorist organizations have exploited social media platforms such as Twitter to spread violent propaganda videos and images and justify their ideology. In addition, terrorist groups use social media platforms to establish one-on-one connections with potential members (Bodine-Baron et al. 2016).
ISIS is one of several terrorist organizations that utilize social media platforms such as Twitter. ISIS conducts an active propaganda campaign on Twitter, highlighting its plans for state-building and violence against groups it deems hostile (Veilleux-Lepage 2016). In particular, the brutal images shared on Twitter by ISIS during its advance in northern Iraq in 2014 drew the global public’s attention, and people worldwide witnessed the inhumane actions of ISIS in northern Iraq (Berger 2014; Richards 2014; Farwell 2014). Recognizing the power and speed of Twitter, ISIS’s official media outlets, fighters and sympathizers actively use the platform, acting as ambassadors for the organization on Twitter (Harbi and Kamsin 2021). ISIS employs videos and pictures of brutal mass killings, beheadings and torture on Twitter to terrorize its adversaries, recruit sympathizers and members, and train recruits worldwide (Chan 2015). Additionally, ISIS utilizes Twitter effectively in Turkey to recruit new members and instill fear among the Turkish people (Tulga 2022b).
Due to ISIS’s active use of Twitter, numerous academic studies in Turkey and worldwide have focused on the Twitter activities of ISIS supporters in recent years. Many of these studies have examined the tweets of ISIS sympathizers and attempted to provide insight into the content of their tweets.
While several studies have highlighted the active use of Twitter by ISIS sympathizers and have analyzed the content of their tweets, there is limited research that compares the tweets of foreign supporters of ISIS with those of supporters from predominantly Muslim countries. Furthermore, few studies have examined the similarities and differences between these tweets. In the case of ISIS’s Turkish sympathizers, prior research has primarily relied on basic text analysis methods, such as the frequency of specific words or qualitative content analysis, rather than providing more in-depth insights.
For this reason, an in-depth and comprehensive study on the tweets of ISIS’s Turkish sympathizers could not be conducted. In this study, I compare the content of the tweets of English-speaking supporters of ISIS and the tweets of Turkish supporters of ISIS. Furthermore, this study provides detailed information on the content of tweets by Turkish supporters of ISIS between 2015 and 2016.
As a result of this study, I have reached significant and informative findings contributing to the existing literature. Firstly, I discovered that most Turkish and English-speaking ISIS supporters had tweeted between zero and 99 times. However, I also found that the Turkish supporter who tweeted the most had tweeted 4,380 times, while the English-speaking supporter who tweeted the most had only tweeted 1,571 times. This suggests that Turkish ISIS sympathizers are more active on Twitter than English-speaking ones, especially among those who tweet frequently.
Secondly, I observed that English-speaking ISIS sympathizers tweeted the most in 2016, whereas Turkish ISIS sympathizers tweeted the most in 2015. This finding highlights the temporal variations in the usage of Twitter by these two groups of ISIS sympathizers.
An analysis of the content of the tweets showed that English-speaking supporters of ISIS frequently tweeted about violence or terrorism, and this trend increased during the period of ISIS’s defeat. In contrast, since ISIS considers Turkey as a recruitment base, most Turkish supporters’ tweets contained religious or justifying content. However, when Turkey became more involved in the fight against ISIS in late 2015, its supporters began incorporating violent or justifying content into their tweets.
Despite the significant findings, the most critical limitation of this study is the lack of access to deactivated accounts among both Turkish and English-speaking supporters of ISIS. Future studies that can address these limitations will contribute to more detailed and comprehensive research in the literature.
Appendix 1: List of Hashtags
#incirlikkapatilsin |
#allahuakbar |
#id |
#incirlikdarbesi |
#abdyikovhilafetikur |
#musul |
#amaqagency |
#aamaq |
#idlib |
#halep |
#halepfethi |
#khilafahnews |
#caravanofmartyrs |
#aleppoisburning |
#freemuslimprisoners |
#wilayatsayna |
#alhayatmediacenter |
#ansaralkhilafahmedia |
#wilayatfallujah |
#greenbird |
#miqdadiyah |
#ISIS |
#IS |
#raqqa |
#dawlah |
#haseke |
#holkampi |
#islamicstate |
#halebinfethi |
#şehidrehberirahmetleanıyoruz |
#bacimadokunma |
#haclikoalisyonunahayir |
#fridayinaleppo |
#totheislamicstate |
#caliphatefollow |
#halifeliktakip |
#allahinaslanimohammedyoussef |
#hilafettakip |
#işid |
#hilafetbiryaşında |
#halepcuması |
#halepkuşatılıyor |
#suriyedemüslümankatliamı |
#hakimiyetallahındır |
#halepeduavakti |
#bagdadi |
#halife |
#AllEyesOnISIS/#alleyesonisis |
Appendix 2: Algorithm (Criteria)
Criteria |
Scholars recommending this criterion |
|
---|---|---|
1 |
ISIS flag as a profile picture, cover picture, or avatar photo |
Özeren, Çubukçu and Cash 2021; Özeren et al. 2018; Conway et al. 2019 |
2 |
Warrior picture as a profile picture, cover picture, or avatar photo |
Conway et al. 2019 |
3 |
Screen name, display name, and handle related to ISIS or Jihad-related words |
Conway et al. 2019; Özeren et al. 2018 |
4 |
ISIS or Jihad-related words in their bio |
Parekh et al. 2018; Conway et al. 2019 |
5 |
Support ISIS-related hashtags |
Bodine-Baron et al. 2016; Özeren et al. 2018; Conway et al. 2019 |
6 |
A location such as Islamic State |
Conway et al. 2019 |
7 |
Dialogue or mention with ISIS supporters whose accounts were suspended or deactivated |
Özeren, Çubukçu and Cash 2021 |
8 |
Anonymous reports about ISIS’s accounts |
Shehabat and Mitew 2018 |
9 |
Positive tweets on ISIS during ISIS attacks; at least like or post two positive tweets about ISIS and its attacks |
Bodine-Baron et al. 2016; Yalçınkaya 2017 |
10 |
Tweeted ISIS slogan |
Bodine-Baron et al. 2016 |
11 |
Random numbers and letters (look again at point six) |
Conway et al. 2019 |
12 |
Following at least three ISIS supporters (look at the sixth point again) |
Özeren et al. 2018; Berger and Morgan 2015 |
13 |
Tweeted, liked or favorite video(s) related to ISIS |
Özeren et al. 2018; Berger and Morgan 2015; Parekh et al. 2018 |
14 |
Tweeted and retweeted ISIS’s media group, such as Al-Hayat Media |
Parekh et al. 2018; Berger and Morgan, 2015 |
I benefited from my unpublished doctoral dissertation titled “New Page in Terrorism: An Explanation of the ‘Soft’ Terrorism Concept with ISIS Case in Turkey” in this study. The tweets of the Turkish sympathizers of ISIS used in this study were also used in the doctoral dissertation.
API, or application programming interface, allows different software programs to communicate with each other.
The reason for beginning the collection of tweets in January 2015 is the increase in the attacks of the anti-ISIS coalition against ISIS during this period and the defeat of ISIS in Kobanî (Kobane) in January 2015 (Colombo and Curini 2022).
See Appendix 1 for the full list of hashtags. I benefited from finding hashtags in the studies of Berger (2014), Vitale and Keagle (2014), Carter, Maher and Neumann (2014) and Özeren et al. (2016). However, over time, I increased the number of hashtags beyond those mentioned by these scholars through my research.
See Appendix 2 for details.
References
Abd-Elaal, A.I., Badr, A.Z. and Mahdi, H.M. (2020). Detecting Violent Radical Accounts on Twitter. International Journal of Advanced Computer Science and Applications 11 (8), pp. 516–522.
Adamic, L.A., and Glance, N. (2005). The Political Blogosphere and the 2004 U.S. Election: Divided They Blog. In Proceedings of the 3rd International Workshop on Link Discovery, New York: ACM, pp. 36–43.
Agarwal, N., Al-Khateeb, S., Galeano, R. and Goolsby, R. (2017). Examining the Use of Botnets and Their Evolution in Propaganda Dissemination. Defence Strategic Communications 2 (1), pp. 87–112.
Ali, M. (2014). ‘Jihad’ in Syria: Fallacies of ISIS’ End Time Prophecies. RSIS Commentaries (149/2014). Singapore: S. Rajaratnam School of International Studies. Retrieved from https://www.rsis.edu.sg/rsis-publication/rsis/jihad-in-syria-fallacies-of-isis-end-time-prophecies/#.XMzFlOhKg2w.
Al-Khateeb, S., and Agarwal, N. (2015, November). Examining Botnet Behaviors for Propaganda Dissemination: A Case Study of ISIL’s Beheading Videos-Based Propaganda. In 2015 IEEE International Conference on Data Mining Workshop, Atlantic City, pp. 51–57, DOI: 10.1109/ICDMW.2015.41.
Al-Rawi, A., and Groshek, J. (2018). Jihadist Propaganda on Social Media: An Examination of ISIS-Related Content on Twitter. International Journal of Cyber Warfare and Terrorism 8 (4), pp. 1–15.
Ashcroft, M., Fisher, A., Kaati, L., Omer, E. and Prucha, N. (2015, September). Detecting Jihadist Messages on Twitter. In 2015 European Intelligence and Security Informatics Conference. Manchester, UK, pp. 161–164, DOI: 10.1109/EISIC.2015.27.
Ayad, M. (2020). The Propaganda Pipeline: The ISIS Fuouaris Upload Network on Facebook. London: Institute for Strategic Dialogue.
Badawy, A., and Ferrara, E. (2018). The Rise of Jihadist Propaganda on Social Networks. Journal of Computational Social Science 1 (2), pp. 453–470.
Barceló, J., and Labzina, E. (2020). Do Islamic State’s Deadly Attacks Disengage, Deter, or Mobilize Supporters? British Journal of Political Science 50 (4), pp. 1539–1559.
Berger, J.M. (2014). How ISIS Games Twitter. The Atlantic, 16 June. Retrieved from: http://www.theatlantic.com/international/archive/2014/06/isis-iraq-twitter-social-media-strategy/372856/.
Berger, J.M. (2015). The Metronome of Apocalyptic Time: Social Media as Carrier Wave for Millenarian Contagion. Perspectives on Terrorism 9 (4), pp. 61–71.
Berger, J.M., and Morgan, J. (2015). The ISIS Twitter Census: Defining and Describing the Population of ISIS Supporters on Twitter. Analysis paper, Brookings Institution. https://www.brookings.edu/wp-content/uploads/2016/06/isis_twitter_census_berger_morgan.pdf.
Berger, J.M., and Perez, H. (2016). The Islamic State’s Diminishing Returns on Twitter: How Suspensions Are Limiting the Social Networks of English-Speaking ISIS Supporters. Occasional paper, George Washington University.
Bodine-Baron, E., Helmus, T.C., Magnuson, M. and Winkelman, Z. (2016). Examining ISIS Support and Opposition Networks on Twitter. Santa Monica, CA: RAND.
Bonikowski, B., and Gidron, N. (2015). The Populist Style in American Politics: Presidential Campaign Discourse, 1952–1996. Social Forces 94 (4), pp. 1605–1654.
Carter, J.A., Maher, S. and Neumann, P.R. (2014). #Greenbirds: Measuring Importance and Influence in Syrian Foreign Fighter Networks. London: International Centre for the Study of Radicalisation. Retrieved from: https://icsr.info/wp-content/uploads/2014/04/ICSR-Report-Greenbirds-Measuring-Importance-and-Influence-in-Syrian-Foreign-Fighter-Networks.pdf.
Ceron, A., Curini, L. and Iacus, S.M. (2019). ISIS at Its Apogee: The Arabic Discourse on Twitter and What We Can Learn from That About ISIS Support and Foreign Fighters. SAGE Open 9 (1), pp. 1–14.
Chan, A. (2015). The Call of ISIS: The Medium and the Message Attracting Southeast Asians. Counter Terrorist Trends and Analyses 7 (4), pp. 4–9.
Chatfield, A.T., Reddick, C.G., and 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. Retrieved from: http://dl.acm.org/citation.cfm?doid=2757401.2757408.
Clark, A. (2016). Individual Well-Being and the Allocation of Time Before and After the Boston Marathon Terrorist Bombing. Economic Journal 130 (631), pp. 2065–2104.
Colombo, M., and Curini, L. (2022). Discussing the Islamic State on Twitter. Zurich: Springer Nature.
Conway, M. (2017). Determining the Role of the Internet in Violent Extremism and Terrorism: Six Suggestions for Progressing Research. Studies in Conflict & Terrorism 40 (1), pp. 77–98.
Conway, M., Khawaja, M., Lakhani, S., Reffin, J., Robertson, A. and Weir, D. (2019). Disrupting Daesh: Measuring Takedown of Online Terrorist Material and Its Impacts. Studies in Conflict & Terrorism 42 (1–2), pp. 141–160.
Dillon, L., Neo, L.S. and Freilich, J.D. (2020). A Comparison of ISIS Foreign Fighters and Supporters Social Media Posts: An Exploratory Mixed-Method Content Analysis. Behavioral Sciences of Terrorism and Political Aggression 12 (4), pp. 268–291.
Ekwenchi, O.C., Ononiwu, C.R. and Efetobor, E. (2016). On the Dialectic of Social Media-Assisted Terrorism: The Case for Content-Based Countermeasures. International Journal of Social Sciences and Humanities Review 6, pp. 11–21.
El-Badawy, E., Comerford, M. and Welby, P. (2015). Inside the Jihadi Mind: Understanding Ideology and Propaganda. London: The Tony Blair Institute for Global Change.
Farwell, J.P. (2014). The Media Strategy of ISIS. Survival 56 (6), pp. 49–55.
Fernandez, M., Asif, M. and Alani, H. (2018, May). Understanding the Roots of Radicalisation on Twitter. In Proceedings of the 10th ACM Conference on Web Science, pp. 1–10, DOI: https://doi.org/10.1145/3201064.3201082.
Ficamos, P., and Liu, Y. (2016). A Topic-Based Approach for Sentiment Analysis on Twitter Data. International Journal of Advanced Computer Science and Applications 7 (12), pp. 201–205.
Finseraas, H., and Listhaug, O. (2013). It Can Happen Here: The Impact of the Mumbai Terror Attacks on Public Opinion in Western Europe. Public Choice 156 (1), pp. 213–228.
Gaikwad, M., Ahirrao, S., Phansalkar, S. and Kotecha, K. (2021). Multi-Ideology ISIS/Jihadist White Supremacist (MIWS) Dataset for Multi-Class Extremism Text Classification. Data 6 (11), pp. 1–15.
Gentry, J. (2016). R-Based Twitter Client. CRAN. Retrieved from: https://cran.r-project.org/web/packages/twitteR/twitteR.pdf.
Gomes, B.G., Holanda, P.H., Couto da Silva, A.P. and Goussevskaia, O. (2017). Profiling ISIS Supporters on Twitter. In Proceedings of the 23rd Brazilian Symposium on Multimedia and the Web. 17 October, Gramado, pp. 457–460. Retrieved from: https://dl.acm.org/doi/10.1145/3126858.3131597, accessed 25 July 2022.
Guthrie, A.R. (2018). #IslamicState: An Analysis of Tweets in Support of ISIS After the November 2015 Attacks in Paris. Master’s thesis, Graduate Faculty of Sociology, Auburn University.
Harbi, A., and Kamsin, A.B. (2021). An Effective Text Classifier Using Machine Learning for Identifying Tweets’ Polarity Concerning Terrorist Connotation. Information Technology and Computer Science 5 (1), pp. 19–29.
Hoffman, Y. (2017). Who Is Afraid of ISIS? ISIS Anxiety and Its Correlates. Stress and Health 34 (1), pp. 84–92.
Kaakinen, M., Oksanen, A., Gadarian, S.K., Solheim, Ø.B., Herreros, F., Winsvold, M.S., … and Steen-Johnsen, K. (2021). Online Hate and Zeitgeist of Fear: A Five-Country Longitudinal Analysis of Hate Exposure and Fear of Terrorism After the Paris Terrorist Attacks in 2015. Political Psychology 42 (6), pp. 1019–1035.
Kemp, S. (2022). Digital 2022: Global Overview Report. Retrieved from https://datareportal.com/reports/digital-2022-global-overview-report.
Klausen, J. (2015). Tweeting the Jihad: Social Media Networks of Western Foreign Fighters in Syria and Iraq. Studies in Conflict & Terrorism 38 (1), pp. 1–22.
Lazer, D., Pentland, A., Adamic, L., Aral, S., Barabási, A.L., Brewer, D. and Van Alstyne, M. (2009). Computational Social Science. Science 323 (5915), pp. 721–723.
Lee, C.S., and Colautti, J.D. (2022). ISIS’s COVID-19 Messaging on Twitter: An Analysis of Tweet Sentiment and Emotions. Crime & Delinquency 10 (5), pp. 1347–1370.
Leite, Â., Ramires, A., Dinis, M.A.P. and Sousa, H.F.P. (2019). Who Is Concerned About Terrorist Attacks? A Religious Profile. Social Sciences 8 (11), pp. 1–16, DOI: https://doi.org/10.3390/socsci8110316
Lieberman, A.V. (2017). Terrorism, the Internet, and Propaganda: A Deadly Combination. Journal of National security Law & Policy 9 (95), pp. 95–124.
Lorenzo-Dus, N., Kinzel, A. and Walker, L. (2018). Representing the West and “Non-Believers” in the Online Jihadist Magazines Dabiq and Inspire. Critical Studies on Terrorism 11 (3), pp. 521–536.
Marin, L. (2016). The Fate of the Data Retention Directive: About Mass Surveillance and Fundamental Rights in the EU Legal Order. London: Edward Elgar.
Mitts, T., Phillips, G. and Walter, B.F. (2022). Studying the Impact of ISIS Propaganda Campaigns. The Journal of Politics 84 (2), pp. 100–116.
Msall, K., and Lary, N. (2022). A Link Between Radicalisation Models and Extremist Propaganda. Global Security: Health, Science and Policy7 (1), pp. 44–50.
Narin, B., and Ayaz, B. (2016). Tweeting for Jihad: ISIS Supporters on Twitter. In 5th International Conference on Conflict, Terrorism, and Society, Non-State Actors in Conflicts: Conspiracies, Myths, and Practices. 12–13 April, Istanbul, pp. 12–13. Retrieved from: https://avesis.hacibayram.edu.tr/yayin/ff643963-8d79-4729-922b-492584f00184/tweeting-for-jihad-isis-supporters-on-twitter
Näsi, M., Tanskanen, M., Kivivuori, J., Haara, P. and Reunanen, E. (2020). Crime News Consumption and Fear of Violence: The Role of Traditional Media, Social Media, and Alternative Information Sources. Crime & Delinquency 67 (4), pp. 574–600.
Neo, L.S., Khader, M., Ang, J., Ong, G. and Tan, E. (2017). Developing an Early Screening Guide for Jihadi Terrorism: A Behavioural Analysis of 30 Terror Attacks. Security Journal 30 (1), pp. 227–246.
Newman, N. (2022). 2022 Digital News Report. Retrieved from https://reutersinstitute.politics.ox.ac.uk/sites/default/files/2022-06/Digital_News-Report_2022.pdf.
Nouh, M., Nurse, J.R.C. and Goldsmith, M. (2019). Understanding the Radical Mind: Identifying Signals to Detect Extremist Content on Twitter. In 2019 IEEE International Conference on Intelligence and Security Informatics (ISI), Shenzhen, China, pp. 98–103, DOI: 10.1109/ISI.2019.8823548.
Oksanen, A., Kaakinen, M., Minkkinen, J., Räsänen, P., Enjolras, B. and Steen-Johnsen, K. (2020). Perceived Societal Fear and Cyberhate After the November 2015 Paris Terrorist Attacks. Terrorism and Political Violence 32 (5), pp. 1047–1066.
Omer, E. (2015). Using Machine Learning to Identify Jihadist Messages on Twitter. Master’s thesis, Uppsala University. Retrieved from http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-260099.
Özeren, S., Çubukçu, S. and Cash, G. (2021). Exposure to Extremist Content and Public Sympathy for ISIS. Studies in Conflict & Terrorism 10 (5), pp. 1–21.
Özeren, S., Hekim, H., Elmas, M.S. and Canbegi, H.I. (2016). ISIS in Cyberspace: Findings from Social Media Research. Ankara, Turkey: Global Policy and Strategy Institute. Retrieved from: https://www.researchgate.net/publication/319164618_ISIS_in_Cyberspace_Findings_From_Social_Media_Research.
Özeren, S., Hekim, H., Elmas, M.S., and Canbegi, H.I. (2018). An Analysis of ISIS Propaganda and Recruitment Activities Targeting the Turkish-Speaking Population. International Annals of Criminology 56 (1–2), pp. 105–121.
Pantucci, R. (2011). What Have We Learned About Lone Wolves from Anders Behring Breivik? Perspectives on Terrorism 5 (5–6), pp. 27–42.
Parekh, D., Amarasingam, A., Dawson, L. and Ruths, D. (2018). Studying Jihadists on Social Media: A Critique of Data Collection Methodologies. Perspectives on Terrorism 12 (3), pp. 5–23.
Redmond, S.A. (2017). Who Views Graphic Media and Why? A Mixed-Methods Study of the ISIS Beheading Videos. Master’s thesis, University of California. Retrieved from: https://escholarship.org/uc/item/7x37b4j6#main.
Rehman, A.U., Iqbal, J., Shakeel, A., Qamar, Z.U. and Rana, P. (2021). Hardy–Weinberg Equilibrium Study of Six Morphogenetic Characters in a Population of Punjab, Pakistan. All Life 13 (1), pp. 213–222.
Richards, A. (2014). Conceptualizing Terrorism. Studies in Conflict Terrorism 37 (3), pp. 213–236.
Seib, P., and Janbek, D.M. (2010). Global Terrorism and New Media: The Post-Al Qaeda Generation. London: Routledge.
Shehabat, A., and Mitew, T. (2018). Black-Boxing the Black Flag: Anonymous Sharing Platforms and ISIS Content Distribution Tactics. Perspectives on Terrorism 12 (1), pp. 81–99.
Shrira, A. (2015). Transmitting the Sum of All Fears: Iranian Nuclear Threat Salience Among Offspring of Holocaust Survivors. Psychological Trauma: Theory, Research, Practice, and Policy 7 (4), pp. 364–371.
Sriram, B., Fuhry, D., Demir, E., Ferhatosmanoglu, H. and Demirbas, M. (2010). Short Text Classification in Twitter to Improve Information Filtering. In Proceedings of the 33rd International ACM SIGIR Conference on Research and Development in Information Retrieval, 17 July, Geneva, pp. 841–842. Retrieved from: https://doi.org/10.1145/1835449.1835643.
Steadman, L.E. (2020). Disengagement and Reconciliation in Conflict-Affected Settings. Washington, DC: United States Institute of Peace. Retrieved from: https://www.jstor.org/stable/pdf/resrep26029.pdf?acceptTC=true&coverpage=false&addFooter=false.
Subramanian, K.R. (2018). Technology and Transformation in Communication. Journal of Advance Research in Electrical & Electronics Engineering 5 (8), pp. 1–13.
Tulga, A.Y. (2022a). Hard and Soft Terrorism Concepts: The Case of ISIS. Pakistan Journal of Terrorism Research 3 (2), pp. 109–132.
Tulga, A.Y. (2022b). New Page in Terrorism: An Explanation of the ‘Soft’ Terrorism Concept with ISIS Case in Turkey. Doctoral Dissertation, Institute of Political Science, National Sun Yat-sen University, Taiwan/Kaohsiung.
Türkoğlu, E. (2017). Küresel bir Terör Örgütü Olarak IŞİD’in Dijital Dergi Kullanımı: Konstantiniyye Üzerine bir İnceleme [ISIS’s use of digital magazines as a global terrorist organization: A study on Konstantiniyye]. Erciyes İletişim Dergisi 5 (1), pp. 162–180.
Veilleux-Lepage, Y. (2016). Paradigmatic Shifts in Jihadism in Cyberspace: The Emerging Role of Unaffiliated Sympathizers in Islamic State’s Social Media Strategy. Contemporary Voices: St Andrews Journal of International Relations 7 (1), pp. 36–51.
Vergani, M., and Bliuc, A.M. (2015). The Evolution of the ISIS’ Language: A Quantitative Analysis of the Language of the First Year of Dabiq Magazine. Sicurezza, Terrorismo e societa 2 (1), pp. 7–20.
Vitale, H.M., and Keagle, J.M. (2014). A Time to Tweet, as Well as a Time to Kill: ISIS’s Projection of Power in Iraq and Syria. Defense Horizons 1 (77), pp. 1–12.
Vujanović, J. (2016). The Digital Caliphate: The Use of Propaganda in Daesh Communication Operations to Legitimise, Radicalise and Recruit. Master’s thesis, University of Vienna.
Waters, G., and Postings, R. (2018). Spiders of the Caliphate: Mapping the Islamic State’s Global Support Network on Facebook. New York, USA: The Counter Extremism Project (CEP). Retrieved from: https://www.counterextremism.com/sites/default/files/Spiders%20of%20the%20Caliphate%20%28May%202018%29.pdf.
Watts, D.J. (2007). A Twenty-First Century Science. Nature 445 (7127), pp. 489–490.
Weinberg, L., and Eubank, W. (2007). A Nested Game: Playing Politics with Terrorism in the United States. New York: Columbia University Press.
Welbers, K., Van Atteveldt, W. and Benoit, K. (2017). Text Analysis in R. Communication Methods and Measures 11 (4), pp. 245–265.
Windsor, L. (2020). The Language of Radicalization: Female Internet Recruitment to Participation in ISIS Activities. Terrorism and Political Violence 32 (3), pp. 506–538.
Winter, C. (2017). Apocalypse, Later: A Longitudinal Study of the Islamic State Brand. Critical Studies in Media Communication 35 (1), pp. 103–121.
Yalçınkaya, H. (2016). Turkey’s Struggle Against the Foreign Terrorist Fighters of DAESH. Perceptions: Journal of International Affairs 21 (1), pp. 27–44.
Yalçınkaya, H. (2017). Foreign Fighters of ISIS and Their Security Threat: Turkey’s Experience (2014–2016). Uluslararasi Iliskiler Dergisi 14 (53), pp. 23–44.
Yüksel, E., and Yıldırım, O. (2016). IŞİD terörü haberleri ve okuyucu değerlendirmeleri [News on ISIS terror and reader reviews]. İNİF E-Dergi 1 (2), 21–37.
Zahrah, F., Nurse, J.R. and Goldsmith, M. (2020, September). # ISIS vs# ActionCountersTerrorism: A Computational Analysis of Extremist and Counter-Extremist Twitter Narratives. In 2020 IEEE European Symposium on Security and Privacy Workshops (EuroS&PW), Genoa, Italy, pp. 438–447, DOI: 10.1109/EuroSPW51379.2020.00065.