A network science approach to identifying conditions for consensus building in social networks

Abstract

This thesis explores how consensus emerges in social networks and online discussions. It introduces a multi-layer model to measure deliberation in Reddit threads and an agent-based model to study social diffusion driven by a committed minority. Findings highlight how discussion structure and actor interactions shape deliberation quality, and how network topology and the placement of the committed minority influence the probability and speed of successful diffusion.