Online Library

As violent extremists continue to surface in online discussion forums, law enforcement agencies search for new ways of uncovering their digital indicators. Researchers have both described and hypothesized a number of ways to detect online traces of potential extremists, yet this area of inquiry remains in its infancy. This study proposes a new search method that, through the analysis of sentiment, identifies the most radical users within online forums. Although this method is applicable to web-forums of any type, the method was evaluated on four Islamic forums containing approximately 1 million posts of its 26,000 unique users. Several characteristics of each user’s postings were examined, including their posting behavior and the content of their posts. The content was analyzed using Parts-Of-Speech tagging, sentiment analysis, and a novel algorithm called ‘Sentiment-based Identification of Radical Authors’, which accounts for a user’s percentile score for average sentiment score, volume of negative posts, severity of negative posts, and duration of negative posts. The results suggest that there is no simple typology that best describes radical users online; however, the method is flexible enough to evaluate several properties of a user’s online activity that can identify radical users on the forums.

Why would a terrorist choose to utilize the Internet rather than the usual methods of assassination, hostage taking, and guerrilla warfare? Conway (2006) identified five major reasons why extremist groups used the Internet: virtual community building, information provision, recruitment, financing, and risk mitigation. Terrorist and extremist organizations can use the Internet to increase their visibility and provide information about the group along with its goals without posing an increased risk to the members. It also allows them to easily ask for, and accept, donations through anonymous financial services such as Dark Coins. These benefits allow these groups to promote awareness of their cause, to convey their message to, and perhaps foster sympathy from a much larger pool of potential supporters and converts (Weimann 2010). Finally, the Internet also provides asynchronous services with global access, with the sender and recipient located at any place, at any time, without the need to link up at a specific time (Wagner 2005). In short, unlike the real world, cyberspace is borderless without limitation, and this makes identification, verification, and attribution a challenge.

This study applies the method of sentiment analysis to the online media released by the Islamic State (IS) in order to distinguish the ways in which IS uses language within their media, and potential ways in which this language differs across various online platforms. The data used for this sentiment analysis consist of transcripts of IS-produced videos, the text of IS-produced online periodical magazines, and social media posts from IS-affiliated Twitter accounts. It was found that the language and discourse utilised by IS in their online media is of a predominantly negative nature, with the language of videos containing the highest concentration of negative sentiment. The words and phrases with the most extreme sentiment values are used as a starting point for the identification of specific narratives that exist within online IS media. The dominant narratives discovered with the aid of sentiment analysis were: 1) the demonstrated strength of the IS, 2) the humiliation of IS enemies, 3) continuous victory, and 4) religious righteousness. Beyond the identification of IS narratives, this study serves to further explore the utility of the sentiment analysis method by applying it to mediums and data that it has not traditionally been applied to, specifically, videos and magazines.

In this study, we select the official websites of four known extremist groups and map the networks of hyperlinked websites forming a virtual community around them. The networks are constructed using a custom-built webcrawler (TENE: Terrorism and Extremism Network Extractor) that searches the HTML of a website for all the hyperlinks inserted directing to other websites (Bouchard et al., 2014). Following all of these hyperlinks out of the initial website of interest produces the network of websites forming a community that is more or less cohesive, more or less extensive, and more or less devoted to the same cause (Bouchard and Westlake, 2016; Westlake and Bouchard, 2016). The extent to which the official website of a group contains many hyperlinks towards external websites may be an indicator of a more active community, and it may be indicative of a more active social movement.

A content analysis was conducted on five different white supremacist online forums to observe the discourse and types of activities occurring within. In addition, web link analysis was conducted on the forums to identify the presence of external links being posted and discussed by members. We found that members used the forums primarily for information provision, recruitment and networking. Based on these results, we discuss the implications that online hate speech have within offline settings, and the affects these activities have on Canadian citizens in light of the recent repeal of section 13 of the Canadian Human Rights Act (1985), the primary tool in Canada with which to deal with hate speech and other activities observed. The insights extracted from this research have provided novel insight into the sentiments and activities of the white supremacist movement online, a relatively unexplored venue of hate speech and propaganda online.