突发事件网络谣言危机预警及模拟仿真研究

Study on the Simulation and Crisis Early-warning of Internet Rumors about Emergencies

[目的/意义]提前进行网络谣言监控和预警是立体化防控网络谣言、增强社会稳定、提高政府执政能力的关键。移动互联网时代,突发事件发生后极易在网络上引起热点舆情、网络危机信息的传播同时为网络谣言的扩散提供良好的土壤,无形中增大了政府部门应对谣言的挑战。[方法/过程]本文采用遗传算法优化BP神经网络构建网络谣言危机预警模型,拟实现对突发事件网络谣言的监控、预警仿真及风险的量化评估。实证分析案例选取天津“8·12”爆炸事故与“和颐酒店女生遇袭事件”,通过计算机对这两起突发事件衍生的网络谣言建立预警模型,并对模拟仿真结果进行验证。[结果/结论]结果表明,遗传算法优化BP神经网络模型在突发事件网络谣言危机预警方面具有较好的适用性,与仅采用BP神经网络模型相比预警的准确性更好。

[Purpose/Significance]Early network rumors monitoring and early warning hold the key to all-encompassing prevention and control of network rumors,enhanced social stability and improved the government's ability of administration.In the era of mobile internet,it tends to arouse hot public opinion.Meanwhile,the dissemination of information about network crisis provide fertile grounds for the spread of network rumor,which increases the challenge of government departments to deal with rumors.[Method/Process]In this paper,genetic algorithm was used to optimize BP neural network to construct the network rumor crisis warning model,and to realize the monitoring,early-warning simulation and quantitative evaluation of the risk.Empirical analysis of the case of Tianjin“8·12”explosion accident and“and the girl attack in the heyi hotel”,we formulated an early warning model based on the two unexpected events through the computer,and the simulation results have been verified.[Result/Conclusion]The results showed that the network model of BP neural network genetic optimized by genetic algorithm had a good applicability in the early warning of the emergency network rumor crisis,and the accuracy of early warning was better than the standard BP neural network model.