Maximizing the performance of advertisements diffusion: A simulation study of the dynamics of viral advertising in social networks

Abstract

In the era of web2.0, marketers are eager to benefit from viral advertising. In this paper we propose a computational network model of viral advertising to examine the maximization of influence within social networks. For our network model we combine both the independent cascade model and the threshold model. We use a spreading threshold to trigger the cascading process, to examine the ways in which advertisements spread across the social network. We also investigate the procedures for choosing an initial set of people to maximize the performance of advertisement spreading. Furthermore, we analyse the impact of network structures on the dynamics of diffusion, and a strategy for combining viral advertising with mass marketing in e-commerce. We also run simulations using a real dataset to check the diffusion of advertisements in an online social network. Ultimately we discovered that a combination of viral advertising and mass marketing is better to diffuse advertisements than either method wholly by itself. Using an optimal algorithm improves diffusion performance, but using ‘degree’ is also an alternative way of choosing initial nodes when the whole structure of network is unknown. Integrating simulations to build a real-time decision support platform will make the diffusion of advertisements more efficient.