Cross-Domain Argumentation Mining: Game-Theoretic Modeling of Persuasion Dynamics in Social Networks

While Sarit Kraus (2016) discusses integrating game theory with agent negotiation, and Castelfranchi & Falcone (2000) examine social trust cognitively, there’s a gap in computationally mapping how argument strategies propagate and mutate within real social networks. This research would adapt game-theoretic concepts—such as Nash equilibria and signaling games—to large-scale argumentation mining on platforms like Twitter or Reddit. By modeling each user as a strategic agent whose persuasion tactics and susceptibility are influenced by their network position and trust relationships, we could uncover new patterns in how arguments spread, polarize, or dissipate. This could inform better moderation tools, targeted interventions to reduce polarization, and deeper theoretical links between social network analysis and argumentation theory.

References:

1. Human-Agent Decision-making: Combining Theory and Practice. Sarit Kraus (2016). Theoretical Aspects of Rationality and Knowledge.
2. Social Trust : A Cognitive Approach. C. Castelfranchi, R. Falcone (2000).