Adaptive immunization in dynamic networks

In recent years, immunization strategies have been developed for stopping epidemics in complex-network-like environments. So far, there exist two limitations in the current propagation models and immunization strategies: (1) the propagation models focus only on the network structure underlying virus propagation and the models are static; (2) the immunization strategies are offline and non-adaptive in nature, i.e., these strategies pre-select and pre-immunize "important" nodes before virus propagation starts. In this paper, we extend an interactive email propagation model in order to observe the effects of human behaviors on virus propagation, and furthermore we propose an adaptive AOC-based immunization strategy for protecting dynamically-evolving email networks. Our experimental results have shown that our strategy as an online strategy can adapt to the dynamic changes (e.g., growth) of networks.