2023 Author: Bryan Walter | [email protected]. Last modified: 2023-05-21 22:24
Scientists have created an algorithm that can identify extremists on social networks even before the posts are published, according to the journal Operations Research. To do this, they analyzed five thousand Twitter accounts.
Extremists associated with the Islamic State (the organization's activities are banned in Russia), as well as other terrorist organizations, usually use social networks for propaganda, intimidation of the population and recruitment of new members. Companies like Facebook and Twitter already block suspicious users using special algorithms, but this happens after the publication of posts. Now, the researchers have created an algorithm that, in the future, will allow extremist accounts to be deactivated immediately after their creation.
Jytte Klausen of Brandeis University in Massachusetts, along with colleagues from MIT, collected data from 5,000 Twitter accounts, which were either ISIS members or people associated with them. Their names were obtained through the media, blogs, law enforcement agencies and think tanks. Researchers collected all the information about users, which was indicated in the profile, and also obtained similar data for their friends and followers. Because of this, the number of profiles in the database increased to 1.3 million. Scientists also analyzed 4.8 million tweets related to extremist accounts: some of them belonged to the users themselves, and some simply mentioned them or contained the corresponding hashtags.
Using statistical simulations of extremist behavior, post analysis, and user data, Clausen's group has created a predictive model that can identify accounts as extremist even if they haven't posted a single microblog. This is due to the fact that the algorithm is based not on the description of posts, but on how a typical extremist profile looks in general. For example, scientists found that 70 percent of the accounts from the new dataset belong to extremists, and the margin of error was only 2 percent.
“Users who engage in some form of online extremism will have very similar behavioral characteristics on social media. They will be associated with a specific set of users who form the extremist group. They will create new, similar accounts after blocking the old ones, and then return to the social network and with a high degree of probability restore connections with the same persons,”explains Clausen.
Scientists sometimes take a non-standard approach to finding extremist accounts and groups on the Internet. For example, a model was recently built that relies on the theory of gelation and describes the growth of extremist groups and showed with its help that extremist groups usually unite people with similar interests, but it is impossible to get rid of network extremism by destroying several "instigators".