基于完全级联传播模型的社区影响最大化

Community Influence Maximizing Based on Comprehensive Cascade Diffuse Model

基于社会网络的社区结构特性和网络中个体间的相互影响,通过引入社区影响最大化的概念,并根据节点间相互影响强度的动态变化,提出一种新的影响传播模型:完全级联传播模型.利用该传播模型进行社区影响最大化研究,在安然邮件数据集上对该传播模型和独立级联模型进行实验对比,结果表明了该模型在社区影响最大化上应用的有效性.

In consideration of the community structure existing in social network and individual＇s interaction, we introduced the concept of community influence maximization. Furthermore, the influence probability among nodes may change due to the dynamic change of the interaction＇ s intensity between the nodes. Thus, in this paper, the authors will propose a new diffuse model named comprehensive cascade model and use this model to study community influence maximization. Through the experiments on Enron email dataset, we compared the model＇ s performance with independent cascade＇ s. The result shows that the model is feasible in the community influence maximization.