Researchers study how terror messages spread on Twitter

Researchers are studying social-media posts of terrorist networks to determine what type of information and propaganda go viral, with an aim to negate their online actions.

As terrorist groups, such as ISIS or al-Qaeda, evolve their social-media skills, researchers are monitoring their advancements and trying to determine how their online actions can be negated.

Arizona State University (ASU) is leading a group project to study what types of information go viral online, and what types of actions or responses can halt the spread of viral information. The team, which includes researchers from the US Military Academy and Britain's University of Exeter, will study information cascades trends marked by people ignoring their own knowledge or information in favour of suggestions from other people's actions as they relate to the social-media posts of terrorist networks.

"The first phase of the project is we are trying to understand what goes viral. The viral (message) is driven by two things - what type of content and what type of network," said Hasan Davulcu, director of ASU's Cognitive Information Processing Systems Lab.

"The right content and the right types of networks are going to resonate and spread and maybe gain new followers," said Davulcu. Once they understand the information cascade, he said they might be able to determine how to counter the viral messages. But, he clarifies, this study will not include developing content to thwart online terrorism.

Rather, the team will be observing what organic information created by social-media users tends to halt terrorists' viral content. "It's the early detection of what works for them and what works for others opposing them," Davulcu said.

The team believes images and videos might be some of the more persuasive ways to create partisan passion. "We are finding pictures to be extremely telling," he said. "In fact, we are going to collect tonnes of photos that circulate online and put them into games so we can figure out what do people understand by the picture," Davulcu said.

Studying the use and relationships of these images could provide a lens into the diffusion of various ideologies. "It is impossible to monitor all of the conversations, so we have to get better at identifying the ones to which we should be paying attention," said Paulo Shakarian, director of the Cyber-Scio Intelligent Systems Lab at ASU.

"This requires embedding psycho-social models in a logic programming framework that can gather and analyse social networks, specific attributes of individuals and their relationships to others," Shakarian said.