Shaping Opinion Dynamics in Social Networks

A networked opinion diffusion process that usually involves extensive spontaneous discussions between connected users, is often propelled by external sources of news or feeds recommended to them. In many applications like marketing design, or product launch, etc., corporations often post curated news or feeds on social media in order to steer the users' opinions in a desired way. We call such scenarios as opinion shaping or opinion control whereby a few select users called control users post opinionated messages to drive the others' opinions to reach a given state. In this paper, we propose SmartShape, an opinion control package that jointly selects the control users, as well as computes the optimum rate of control messages, thereby driving the networked opinion dynamics to the desired direction. Furthermore, our proposal also includes a robust shaping suit which makes our control framework resilient to stochastic fluctuations of opinion dynamics, originating from several sources of randomness. Experiments on several synthetic and real datasets gathered from Twitter, show that SmartShape can accurately determine the quality of a set of control users as well as shape the opinion dynamics more effectively than several baselines.