异质信息网络中基于邻接熵的影响力最大化

Influence maximization based on adjacency entropy in heterogeneous information networks

影响力最大化是研究如何在社会网络中寻找小部分最具影响力的节点作为信息扩散源,使信息在网络中传播范围最大的问题。已有相关研究大多只是针对同质信息网络,但现实中的社会网络是包含了多种对象类型和对象之间多种关系类型的异质网络,因此提出了基于元路径的邻接信息熵(MPAIE)模型,以及基于元图的邻接信息熵(MGAIE)模型来模拟异质信息网络中的社会影响。通过设置元路径或元图,该方法可以灵活地整合异质网络中的结构和语义信息,对节点的影响力做出度量,并在两个真实数据集上验证了MPAIE及MGAIE模型的有效性。

Influence maximization is to study how to find a small number of the most influential nodes in the social network as the source of information diffusion,so as to maximize the scope of information dissemination in the network.Most of the existing researches only focus on homogeneous information networks,but in reality,social networks are heterogeneous networks with multiple object types and multiple relationship types between objects.This paper proposed a meta path based adjacency information entropy(MPAIE)model and a meta graph based adjacency information entropy(MGAIE)model to simulate the social impact in heterogeneous information networks.By setting meta path or meta graph,the proposed method can flexibly integrate the structural and semantic information in heterogeneous networks,and measured the influence of nodes.The effectiveness of MPAIE and MGAIE models is verified on two real datasets.