With high-speed communication and information sharing in social networks, the effective immunity to specific false information would markedly reduce the loss brought by the spreading of false information. To date, most studies only focus on the immunity of positive relationships for information spreading between individuals. However, negative relationships also exist in social networks and might have a strong influence on information spreading. In this study, three strategies of structural immunization and a heuristic strategy are proposed for signed social networks with both positive and negative relationships, which can effectively control the spread of false information by the users with opposing attitudes. After selecting the information spreading model suitable for signed networks, the influence of structural immunization strategies on spreading ranges is explored and shows different phase transitions which indicate that both positive and negative edges play a significant role in immune processes. Then, the results of two evaluation indices showed that the three proposed strategies had better immunity than the state-of-the-art approaches. In addition, to decrease algorithm complexity and achieve better performance, the results of the proposed structural strategies are selected to be the initial values of a heuristic strategy driven by a genetic algorithm. This study contributes to a deeper understanding of the role of negative relationships in information immunity and promotes the application of signed networks in information spreading and control.

随着社交网络的高速传播和信息共享，对特定虚假信息的有效免疫将显着降低虚假信息传播带来的损失。迄今为止，大多数研究只关注积极关系对个体之间信息传播的免疫力。然而，负面关系也存在于社交网络中，可能对信息传播产生强烈影响。在这项研究中，针对具有正负关系的签名社交网络，提出了结构免疫和启发式策略三种策略，可以有效地控制具有相反态度的用户传播虚假信息。在选择适合有符号网络的信息传播模型后，探讨了结构免疫策略对传播范围的影响，并显示了不同的相变，这表明正边缘和负边缘在免疫过程中都起着重要作用。然后，两个评估指标的结果表明，三种提出的策略比最先进的方法具有更好的免疫性。此外，为了降低算法复杂度并获得更好的性能，所提出的结构策略的结果被选择为由遗传算法驱动的启发式策略的初始值。本研究有助于更深入地理解负向关系在信息免疫中的作用，促进签名网络在信息传播和控制中的应用。