PNNL Data Scientists Co-Author Book Chapter about User Engagement with Deceptive News on Social Media
PNNL data scientists Maria Glenski and Svitlana Volkova are studying who engages with deceptive news, how quickly they engage, and what comments they leave publicly
PNNL data scientists Maria Glenski and Svitlana Volkova have contributed a chapter to a book titled Disinformation, Misinformation, and Fake News in Social Media: Emerging Research Challenges and Opportunities. The book, edited by researchers at Arizona State University and Penn State University, will be published in June by Springer.
Glenski and Volkova’s chapter, “User Engagement with Digital Deception,” ties together their recent work exploring how users engage with social media posts on Twitter and Reddit that are linked to deceptive news sources. These news sources have been previously identified as spreading clickbait, conspiracy theories, or disinformation. Srijan Kumar, a computer scientist at the Georgia Institute of Technology, also co-authored the chapter.
The chapter describes how the researchers use computational social science to study interactions and engagement, along with natural language processing to extract reactions and infer personal characteristics from posts and comments.
“The analysis provides information on the fundamentals of user engagement with deceptive news: who engages, how quickly they engage, and what feedback they provide that others can see publicly,” Glenski says.
The researchers found that users express more appreciation for posts that link to deceptive news sources compared to trustworthy news, among other results.
“The idea that humans will identify and surface deceptive news on social media is not necessarily true,” Glenski says. “Our results provide context for understanding how to leverage the wisdom of social media users to defend against the spread of misinformation locally, nationally, and globally.”