Phantom cascades: The effect of hidden nodes on information diffusion

Abstract

Research on information di usion generally assumes complete knowledge of the underlying network. However, in the presence of factors such as increasing pri- vacy 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 e ect of hidden network structure on information di usion processes. We characterise information cascades through activation paths traversing vis- ible and hidden parts of the network. We quantify di usion estimation error while varying the amount of hidden structure in ve empirical and synthetic network datasets and demonstrate the e ect of topological properties on this error. Finally, we suggest practical recommendations for practitioners and pro- pose a model to predict the cascade size with minimal information regarding the underlying network.