【摘要】 随着电子设备的日益普及和信息扩散的便利性，在线社交网络为各种负面信息的传播提供了高效的媒介.谣言是社交媒体上负面信息的突出形式之一，会引发社会动荡，造成经济损失，因此，快速有效地抑制谣言传播成为当前社交网络研究领域中的一个热点.提出一种有效的谣言抑制传播方法，从网络中选取多个正种子节点来传播真相，抑制谣言的传播.首先采用竞争性独立级联（Conpetitive Independent Cascade,CIC）模型来同时传播谣言和真相；其次，提出一种基于标签传播的社区检测算法对社交网络进行分解，并为各个社区分配正种子节点预算；最后，创新地提出节点强度来衡量网络中节点的重要性，并利用节点强度在各个社区中选取抑制谣言传播的初始正种子集.实验证明，该方法能达到与贪婪算法相匹配的抑制效果，且运行时间比贪婪算法快三个数量级.更多还原

【Abstract】 With the increasing popularity of electronic devices and the convenience of information diffusion,online social networks provide an efficient medium for the propagation of various negative information. Rumors are one of the prominent forms of negative information on social media,which trigger social unrest and cause economic losses. Therefore,how to quickly and effectively block the spread of rumors has become a hot spot in current social network research field. In this paper,we present an effective method for blocking rumor propagation,which selects multiple positive seed nodes from the network to spread the truth to block rumor propagation. Firstly,we adopt a Competitive Independent Cascade(CIC) model to propagate rumors and truth simultaneously. Secondly,we propose a community detection method based on label propagation to decompose social networks and allocated positive seed node budgets to each community. Finally,we propose a novel node strength to measure the importance of nodes in the network and use it to select the initial positive seed set which blocks the spread of rumors in each community. Experimental results show that the proposed method achieves the same blocking effect as the Greedy algorithm,while the running time is three orders of magnitude faster than the Greedy algorithm.更多还原