Analysis of Influential Features for Information Diffusion

We analyze information diffusion focusing on network structures. We propose a network growth model that can produce networks with the required features for analysis and we perform validation experimentation using Twitter and Facebook networks. The proposed model produces networks having features calculated from these networks with high accuracy. Using this proposed model, we produce several networks exhibiting various features. We simulate information diffusion on these networks using an asynchronous independent cascade (ASIC) model and calculate the average of influence degree (AID). The AID increases more when there are several small hub nodes than when there area few large hub nodes and when there are many links between the nodes, except for the hub nodes.