面向社交网络用户建模的异质网络研究

Heterogeneous Networks Based Social Network User Modeling

用户建模已经引起了学术界和工业界的广泛关注。现有的方法大多侧重于将用户间的人际关系融入社区,而用户生成的内容(如帖子)却没有得到很好的研究。为此,该文通过实际舆情传播相关的分析,表明在舆情传播过程中对用户属性进行研究的重要作用,并且提出了用户资料数据的筛选方法。同时,该文提出了一种通过异构多质心图池为用户捕获更多不同社区特征的建模。该方法首先构造了一个由用户和关键字组成的异质图,并在其上采用了一个异质图神经网络。为了方便用户建模的图表示,提出了一种多质心图池化机制,将多质心的集群特征融入表示学习中。在三个基准数据集上的大量实验表明了该方法的有效性。

User modeling has attracted great attention in both academia and industry.Most of the existing approaches are focused on incorporating the personal relationships in communities,while the users’generated content such as posts is not well utilized.In this paper,we show that the research on user attributes plays an important role in the process of public opinion dissemination,and propose a screening method of user data.Meanwhile,we propose an approach to capture more diverse community characteristics via heterogeneous multi-centroid graph pooling for user modeling.Specifically,we construct a heterogeneous graph where the nodes consist of both users and keywords and adopt a heterogeneous GCN on it.To facilitate the graph representation for user modeling,we then propose a multi-centroid graph pooling mechanism,which incorporates the affiliated group features with multiple centroids into representation learning.Extensive experiments on three benchmark datasets show the effectiveness of our proposed approach.