The aim, researchers said, is to help organizations, including governments and emergency services, to respond more effectively to events or situations being reported on online social networks.
The project at the University of Sheffield grew from research based on the use of social media during the London riots in 2011.
"There was a suggestion after the 2011 riots that social networks should have been shut down, to prevent the rioters using them to organize," lead researcher Kalina Bontcheva told the BBC.
"But social networks also provide useful information. The problem is that it all happens so fast and we can't quickly sort truth from lies."
The proposed system will analyze data including posts on Twitter, comments in healthcare forums, public comments on Facebook and other online activity in an effort to identify four types of online rumors: Speculation, controversy, misinformation -- untrue rumors spread unwittingly -- and disinformation, false information knowingly spread with malicious intent.
The system will analyze the sources of information to assess their authority, the researchers said, and examine an online account's history or background to try to determine if it has been created just to spread rumors.