Defense Strategy for Social Network Fake News Under Non-Real-Time Data Collection and Disjointed Decision Making

For the multi online social networks (MOSNs) rumor suppression problem, a suppression strategy with non-real-time data collection (NDC) and disjointed decision making (DDM) is designed. First, based on the characteristics of rumors spreading online, a model for the presence of rumor propagation in MOSNs is constructed, where decision makers can influence the number of people who resist rumors, and uncertainty is taken into account to characterize the complexity of the crowd when communicating. Specifically, the constructed model provides a novel analytical perspective to curb rumor propagation through external forces, and the unknowns considered give the model error tolerance and can be further explored. Second, for the problem that decision makers are unable to collect data from various groups in MOSNs in real-time due to the high real-time collection cost, the designed NDC algorithm can ensure that within the calculated upper boundary, decision makers can use non-real-time data to make decisions to suppress rumor spreading, which improves the ability of the algorithm to accept non-real-time data. Moreover, the DDM framework is designed to improve the tolerance of the decision maker in passing the strategy to the executor in non-real-time, which can ensure that the strategy executed by the executor can complete the rumor suppression when the strategy deviation is not larger than the given upper bound. Finally, the effectiveness of the proposed method is verified by Lyapunov theory and simulation results.