Virginia Tech researchers studied the crowdsourced investigation on Jan. 6,
Credits: Photo by Olivia Coleman for Virginia Tech.
How did online sleuths manage to replace wanted posters?
Researchers at Virginia Tech’s Department of Computer Science answered this question by studying a crowdsourced online investigation following the January 6, 2021 uprising in the US Capitol.
“These online communities can provide real value, if managed in the right way,” said Kurt Luther, professor of computer science and history. “Nowadays it’s not just a digital witch hunt creating fake identifications. They can really make a powerful contribution to an ongoing investigation by getting their picture in front of people, which is what law enforcement has been trying to do ever since those wanted posters 200 years ago.
Led by Tianjiao “Joey” Yu, Ph.D. Luther co-advising students, the group analyzed the online community of Sedition Hunters to understand how the group successfully used Twitter to help law enforcement and prosecutors – including the FBI and the Department of Justice – identify suspects who took part in the rebellion. Their work has been cited in numerous arrests and legal proceedings.
Yu is scheduled to present the resulting paper, “Sedition Hunters: A Quantitative Study of the Crowdsourced Investigation into the 2021 US Capitol Attack,” at the annual international meeting, the ACM Web Conference, May 2 in Austin, Texas.
“Often I don’t clearly see the real-life application of my research,” says Yu, whose research interests include group intelligence, natural language processing and linguistics. “But with this, I feel like I’ve really solved a problem and I’ve really made a contribution.”
Yu and Luther collaborated on the project with Ismini Lourentzou, an assistant professor of computer science, and Sukrit Venkatagiri, a postdoctoral researcher at the University of Washington. Venkatagiri is a Hokie double, having earned a master’s degree and a Ph.D. in computer science in 2020 and 2022 respectively. Luther is his Ph.D advisor when this part of the project is finalized.
The research team found that the Stalking Hunters generally adhered to ethical investigation guidelines, managed to collect large amounts of data to create multi-angle composite images, and used creative nicknames as hashtags to collectively refer to strangers and reach a wider audience across the globe. Twitter.
“They’re really focused on what they can do that’s beneficial and not harmful, and I think that’s one of the reasons they’ve been able to keep that going for more than two years,” Luther said.
Crowdsourcing information as a crime-solving tactic is far from a new concept, according to Luther, who directs Virginia Tech’s Crowd Intelligence Lab in Arlington.
“In a way, the first manhunt followed the assassination of Abraham Lincoln because it was the first time photos were actually used to try to identify some of the conspirators,” he said.
More recently, failed attempts to use online crowdsourcing following the 2013 Boston Marathon bombing cast a negative light on the method.
Supported by a National Science Foundation grant, Luther and the Crowd Intelligence Lab began studying the use of crowdsourcing in active investigations in 2017. He says the events of January 6, 2021 provide a unique setting to study.
“It is the most widely documented crime in history, according to Time magazine,” Luther said. “It’s kind of a perfect storm of important events that happened and the huge amount of evidence documented in photos and videos online that are shared publicly, so it creates an opportunity for people who want to help law enforcement to be able to contribute.”
Luther said a key element of the community’s success was law enforcement’s interest in its findings, despite the mistakes online detectives made after the Boston Marathon bombing. “They seem very receptive to what the community can offer,” Luther said.
According to the team’s research, the Sedition Hunters account had more than 66,000 followers which generated over 320,000 tweets with over 1 million retweets between January 12, 2021 and January 12, 2022. Despite its large audience, community content is dominated by a small number of highly influential accounts. . The majority of tweets ask other users to share content or share information regarding news or current events.
The research team found that, early on, Sedition Hunters ran into trouble when some Twitter users identified someone in a tweet incorrectly. Luther said he was impressed how quickly a few power users were able to change most of the behavior.
“They said, ‘Hey, no more mentioning suspects in public. If you think you know who it is, send it to the FBI,’” Luther said. “And for the most part, that’s what the community does. The community is growing.”
The Sedition Hunters proceeded to develop pseudonyms for the unidentified suspects, which they used as searchable hashtags. Luther credits the community’s creativity, using humor to develop names like #LittleRedRioter and #SlimMcTreason, as part of the reason for the success of the effort.
He also said he was surprised by the ability of the public to find and use open source camera footage to create composite images of suspects. This helps overcome obstacles in the first images that are collected, such as the desired person wearing a mask or being captured at a usable angle.
Yu said he was impressed by the ability of the Sedition Hunters collective to take the content generated on Twitter and develop other websites with high-tech features. He specifically points to the development of timelines and maps linked to various video sources as impressive investigative tools.
“It was a complete shock,” said Yu. “They built several websites and it was really interesting to see people interacting with all these fancy tools.”
While overwhelmingly positive, Yu said he saw areas for improvement in the future. This includes becoming more centralized, finding better ways to track the progress of each investigation, and more monitoring of behavior within groups.
Overall, Luther says the Sedition Hunters have shown they have the potential to provide great benefits to law enforcement and society at large.
“This paper offers a lot of promise for how law enforcement and the public can work together,” Luther said. “Officials get a more diverse subset of the population to see suspects and ultimately help solve crimes in an ethical and more effective manner.”