The rate at which the Islamic State (ISIS) continues to recruit people of all walks of life, including women and Westerners, has continued to baffle the world's leading powers.
Recent statistics show that as many as 200,000 foreign fighters have joined the group's ranks since it seized large swaths of land last summer; as many as 4,000 of these have come from Europe alone.
Since the stereotypical figure of the marginalized jihadist has often been at odds with reality - those who join tend to be better educated, wealthier and more successful than the average -, further insight is needed into the profiles of people who are likely to join or support the militant group.
A group of computer scientists in Qatar claim that they have done just that. According to MIT Technology Review, Walid Magdy and colleagues at the Qatar Computing Research Institute in Doha devised a method for predicting who is likely to support or oppose the Islamic State (ISIS) in the future, based on an analysis of their tweets.
In the study published this month, the group studied tweets in Arabic generated by people who support ISIS and those who oppose it to determine what characteristic each group has in common. They conducted an exhaustive search of their pre-ISIS tweets to reveal any common traits that might be determining factors in their later support or opposition of the group.
The group collected more than 3.1 million Arabic tweets with the word "ISIS" created by more than 250,000 users in a three-month span in 2014. Of these users, 165,000 of them had been tweeting before the appearance of the group.
To identify the difference between those supporting and opposing ISIS, they asked a native Arabic speaker to judge the level of polarization of a random sample of 1,000 tweets.
The group found that those who tended to support ISIS always used its full name, while those who opposed it used the abbreviation. By examining the hashtags people used, they also found insight into the origin of support or opposition.
"Looking at discriminating hashtags suggested that a major source of support for ISIS stems from frustration with the missteps of the Arab Spring," said Magdy. "As for opposition to ISIS, it is linked with support for other rebel groups, mostly in Syria, that have been targeted by ISIS, support for existing Middle Eastern regimes, and Shia sectarianism."
They also noticed that support for or opposition to the group varied with time, in accordance with the latest news.
"Anti-ISIS tweets generally peaked when news of ISIS human rights violations emerged such as the killing of hostage, accounts of torture, or reports of the enslavement of Yazidi women," they said. "On the other hand, pro-ISIS tweets generally peaked in conjunction with the release of propaganda videos and major military achievements."
The group trained a machine learning algorithm to identify users of both types and said it was able to classify other users as likely to become pro- or anti-ISIS with high accuracy. "We train a classifier that can predict future support or opposition of ISIS with 87 percent accuracy," they said.