Phantom cascades: The effect of hidden nodes on information diffusion

Abstract

Research on information diffusion generally assumes complete knowledge of the underlying network. However, in the presence of factors such as increasing privacy awareness, restrictions on application programming interfaces (APIs) and sampling strategies, this assumption rarely holds in the real world which in turn leads to an underestimation of the size of information cascades. In this work we study the effect of hidden network structure on information diffusion processes. We characterise information cascades through activation paths traversing visible and hidden parts of the network. We quantify diffusion estimation error while varying the amount of hidden structure in five empirical and synthetic network datasets and demonstrate the effect of topological properties on this error. Finally, we suggest practical recommendations for practitioners and propose a model to predict the cascade size with minimal information regarding the underlying network.

Information Cascades Of Economic Thinking: The Formation Experience Of Behavioral Media Patterns

“…As Bikhchandani et al (1998) note, information cascades can influence the behavior of economic subjects. Meanwhile, Mahdi (2017) is considering information cascades as a part of complex networks, when Belák et al (2016) and Sattari & Zamanifar (2018) focus on the effect of hidden nodes on information diffusion.…”

Joint Inference of Diffusion and Structure in Partially Observed Social Networks Using Coupled Matrix Factorization

“…On the other hand, a cascade has a limited lifespan and spreads over hours to days, so the sequence from beginning to end of a cascade is usually fixed within a limited data collection interval. (2) We are not interested in hidden nodes that do not have any global or local signals in structure or diffusion data. However, we can somewhat handle users with private accounts.…”

“…(3) We consider that the source and mediators of an information dissemination process are internal factors of the network, and any external agents are ignored. As mentioned before, incomplete data in the collected set from a social network is inevitable, and missing data can significantly affect the difference between the output of methods and what actually happens in the real world [2].…”

Modeling Random Forwarding Actions for Information Diffusion over Mobile Social Networks

“…The second category focuses on analyzing the effect of network nodes (users) on information diffusion using different mathematical models [22][23][24][25][26][27][28]. Kimura et al [22] considered the optimization problem of extracting the most influential nodes over a social network.…”

“…More recently, Saito et al [27] proposed an efficient method to find a new kind of influential nodes (supermediators) over a social network and characterized the properties of supermediators. From another perspective, Belák et al [28] studied the effect of hidden nodes on information diffusion and characterized information cascades.…”