社交网络中，研究谣言和辟谣在传播过程中的规律，有助于把握谣言信息传播趋势从而调节舆论走向。本文提出了一种基于情感因素和信任机制的S/E-F(D)-I（Susceptible/Educated-Forwarding(Defuting)-Immune）谣言/辟谣两阶段的传播动力学模型。模型基于传统的SFI（Susceptible-Forwarding-Immune）模型，分别研究谣言信息和相应辟谣信息的传播过程。在第一阶段谣言信息传播过程中考虑情感因素，进一步考察不同极性情感的动态演化过程；在第二阶段辟谣信息传播过程中考虑信任机制，进一步考察网民信任危机对辟谣信息传播的影响。两阶段模型均加入了受教育人群（E）探究辟谣信息相关教育程度及个人认知水平对信息传播的影响。基于已构建的S/E-F(D)-I动力学模型对新浪微博真实案例事件进行了数值拟合，刻画了真实舆情传播过程并验证了已有模型的有效性和合理性，通过参数敏感性分析挖掘影响谣言传播的重要因素，为制定谣言应对决策策略提供了支持。

In social networks, studying the patterns of rumors and counter-rumor in propagations process can help to predict and understand the trend of rumor spreading and design optional inferrentions to guide the public option. Here, we proposed a novel emotion-based S/E-F(D)-I (Susceptible/Educated-Forwarding (Defuting)-Immune) two-stage rumor/counter-rumor propagation dynamics model. Our approach was based on the traditonal SFI (Susceptible-Forwarding-Immune) model, which was used to investigate the propagation process of rumor information and the corresponding counter-rumor information separately. However, our two stage model allowed us to examine the dynamic process of different polaritiese emotions by considering emotional factors in the rumor propagation at the first stage, and then incorporate trust mechanism to quantify the impact of confidence crisis on the dissemination of counter-rumor information at the second stage. Our model also incorparated the educated population (E) at both the 1st and 2nd stage to explore the influence of education and personal knowledge level on information dissemination in the relevant populations. We applied this constructed S/E-F(D)-I dynamics model to numerically fit and simulate some real case events on Sina Weibo to recover the real public opinion propagation process, and to validate the model. We perform sensitivity analysis to determine significant factors affecting rumor propagation in order to suppport the formulation of decision-making strategies for rumor management.