Stochastic Analysis of Multiplex Boolean Networks for Understanding Epidemic Propagation

Abstract

Many large systems are not isolated but rather an integration of several parallel systems, referred to as multiplex networks. Aiming to improve the evaluation efficiency of a simulation-based approach, stochastic computational models are proposed for multiplex Boolean networks with non-Bernoulli sequences encoding signal probabilities. Then, the epidemic spreading process consisting of awareness diffusion on the virtual contact layer and epidemic spreading via physical contacts, is further considered. Given the impacts of nodes in the virtual contact layer, several benchmarks are used to test the average infection probability. The computational results indicate that a node with a larger spreading degree is likely to be an effective target for affecting the average infection probability, which extends the scope of existing observations, although the network topology also plays an important role in determining the infection effect.