Sequential Seeding in Complex Networks: Trading Speed for Coverage

Information spreading in complex networks is often modeled as disseminating information with certain probability from nodes that possess it to their neighbors that do not. Information cascades are triggered when the activation of a set of initial nodes – seeds – results in diffusion to large number of nodes. Here, several novel approaches replacing the commonly used strategy of initiating all seeds at once with a sequence of stages for seed initiation are introduced. Sequential seeding is compared with a single stage approach using real complex networks and applying various seed selection methods and diffusion parameters. The experimental results indicate that sequential seeding strategies deliver better results than single stage seeding in about 90% of cases. Sequential strategies at later stages avoid seeding nodes that are already activated by diffusion active between stages. The gain arises when a saved seed is allocated to a node difficult to reach via diffusion. The new seed may activate network parts not accessible by traditional single stage seeding approaches. Longer seeding sequences tend to activatemore nodes but they also extend the duration of diffusion. Various variants of sequential seeding resolve differently the trade-off between the coverage and speed of diffusion.