Public authority control strategy for opinion evolution in social networks.

Xi Chen,Xiangyun Xiong,Minghong Zhang,Wei Li
DOI: https://doi.org/10.1063/1.4960121
2016-08-03
Abstract:This paper addresses the need to deal with and control public opinion and rumors. Existing strategies to control public opinion include degree, random, and adaptive bridge control strategies. In this paper, we use the HK model to present a public opinion control strategy based on public authority (PA). This means utilizing the influence of expert or high authority individuals whose opinions we control to obtain the optimum effect in the shortest time possible and thus reach a consensus of public opinion. Public authority (PA) is only influenced by individuals' attributes (age, economic status, and education level) and not their degree distribution; hence, in this paper, we assume that PA complies with two types of public authority distribution (normal and power-law). According to the proposed control strategy, our experiment is based on random, degree, and public authority control strategies in three different social networks (small-world, scale-free, and random) and we compare and analyze the strategies in terms of convergence time (T), final number of controlled agents (C), and comprehensive efficiency (E). We find that different network topologies and the distribution of the PA in the network can influence the final controlling effect. While the effect of PA strategy differs in different network topology structures, all structures achieve comprehensive efficiency with any kind of public authority distribution in any network. Our findings are consistent with several current sociological phenomena and show that in the process of public opinion/rumor control, considerable attention should be paid to high authority individuals.
Medicine
What problem does this paper attempt to address?