Detail:We investigate the contagion of ideas in the online world and their effect in persuading groups of individuals to take action in the real world. We find a significant gap in current approaches that do not fully account for the nuances of online group behavior in addition to largely excluding the characteristics and intentions of the receiver of messages in interpersonal communication. Several differences exist between face-to-face and online interactions that are crucial and require theoretical and empirical investigation in the online context. To address limitations in the current state-of-the-art approaches, we develop an integrative model of group behavior and persuasion that will advance computational modeling of human behavior in online settings. Our overall approach starts from repositories of online data that are specifically targeted for their explanatory use. We leverage cutting-edge social science research and theories to derive large-scale annotated datasets and utilize these to build our computational models using novel methods in artificial intelligence, namely recurrent neural networks.

Publications from this project

C43, C45, C49