CRII: HCC: Discovery, generation, and impact of social media user framing of current events

People often turn to online sources for their news about current events, including natural disasters, disease outbreaks, and more. Online news sources often present events from a particular point of view (or "frame"), which explains why things happen, what is important about them, and how they fit into a bigger story. When people read news stories and share them in social media, the way they share and comment on them may support the original story's frame, or they may try to introduce their own reframing (spin).

This project's goal is to understand how the framing of stories changes as they are shared and to develop tools that help social media users better understand the framing of the posts they view. This has the potential to help people think more critically about the content they view online.

To achieve the goals of the project, the research team will develop a new dataset containing news stories and social media posts that reshare the stories, along with user-written comments. Student researchers will be trained to label posts based on issues, sentiment, and other factors that indicate how the news stories and comments attempt to frame the events being shared.

The research team will build AI models using the labeled dataset, then use the models to label a much larger dataset of other articles and posts. This expanded dataset will be analyzed using natural language processing and computational social science methods to better understand how the framing of events changes as stories are reshared across the internet.

The research team will also explore how AI models can be used to generate posts with a specific framing based on the same underlying information. Finally, the team will investigate how posts written by people and by AI-enabled systems with different framings of the same events might affect how people understand them.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.