LITTLE ROCK, Ark. – The University of Arkansas at Little Rock has been awarded a contract from the U.S. Office of Naval Research to develop a web-based tool and app to detect bot accounts on social media.
“We have found that these bots have evolved drastically. They have started from very naive behaviors to highly sophisticated coordinated behaviors,” Dr. Agarwal, a professor of Information Science for at UA-Little Rock.
Dr. Agarwal and his team will receive funding of $75,000 out of the $140,000 contract that will go to further their research on bot detection and hopefully find a solution to “bad bots.”
“For the last 7 years I have studied the evolution of bots behaviors and their coordination tactics,” explains Agarwal.
Bots are fake social media accounts designed to mimic human behavior and influence opinion on social media, says Agarwal.
He says bots can cause major havoc on the public by sharing misinformation or in some cases creating virtual disagreements between users.
“We’ve seen malicious bots can have a big impact and can affect beliefs and behaviors,” he says.
While bots can be used for everyday tasks, like indexing search engines, malicious bots can spread fake news.
“Many of us have encountered bots at some point and might have used them,” says the Information Science professor. “Siri and Alexa are forms of bots that can do simple tasks that are requested of them.”
Agarwal is also the director of the Collaboratorium for Social Media and Online Behavioral Studies (COSMOS). He says people can use the COSMOS website to report misinformation.
“We plan on using an algorithm that performs bot detection, characterization, and threat assessment using Twitter data,” he explains.
Overall, Agarwal says he hopes the work that he and his students will do can be a benefit to everyone.
For more information on this project, click here.