Fil Menczer, professor of computer science and informatics at the School of Informatics, Computing, and Engineering, is part of a group that has been awarded a $1.2 million grant from the Defense Advanced Research Projects Agency (DARPA) to study how and at what rate information spreads in a global information environment.
The project, “COSINE: Cognitive Online Simulation of Information Network Environments,” also involves Professor of Informatics Alessandro Flammini and Assistant Professor of Informatics and Computing YY Ahn. SICE is a subcontractor for the effort, which is a collaboration with the USC Information Sciences Institute (ISI) and the University of Notre Dame.
“Given the recent surge in coordinated campaigns to manipulate public opinion through deceptive social bots and fake news around the world, the goals of this research are important and urgent,” Menczer said. “Other potential applications spaces include sharing correct information with victims during natural disasters, and measuring the effects of online censorship by oppressive regimes.”
The COSINE project is part of a larger DARPA program called Computational Simulation of Online Social Behavior, or SocialSim. It aims to develop innovative technologies for high-fidelity computational simulation of online social behavior focusing specifically on information spread and evolution. High-fidelity (i.e., accurate, at-scale) computational simulation of the spread and evolution of online information supports efforts to analyze strategic disinformation campaigns by adversaries, and could potentially contribute to other critical missions in the online information domain.
Researchers from SICE will be pursuing three main goals in the project. First, they hope to develop a cognitive network model that describes the consistencies and inconsistencies of a belief system while also establishing the simulation mechanisms that determine how a given idea is best accepted and incorporated by an agent to fit into preexisting beliefs. The group also plans to create analytical techniques to enable big-picture descriptions of large sets of agents with similar characteristics. Those models will then be tested for scalability and accuracy, which will help develop reliable techniques to fit the resulting model parameters. Finally, the project plans to develop generative, realistic models of online social networks that will incorporate a number of observed correlations among connected agents.
“We have been doing research in this area for several years, especially focusing on the diffusion of information on social media, using computational models and simulations to study the factors that influence the viral spread of information,” Menczer said. “High-fidelity computational simulation of online information spread could lead to a deeper understanding of how nation-states and non-state actors can use global information platforms to further their goals. Our research may have broad societal benefits for the development of safeguards that will protect the free exchange of information among citizens.”
The collaboration with USC/ISI will include Emilio Ferrara, a former postdoctoral fellow at SICE.
“We’re excited to be part of this collaboration that could be so critical to understanding how information evolves over time on a global scale,” said Raj Acharya, dean of SICE. “Distilling raw data down to a usable form that can be turned into knowledge is at the heart of many of our programs at SICE, and I’m thrilled we can play a critical role in this research.”
The SocialSim program will run in two phases over the next four years, and IU is expected to take part in the full duration of the program.
For more information on the program, visit DARPA’s website.