Dynamic heterogeneous networks describe the temporal evolution of interactions among nodes and edges of different types. While there is a rich literature on finding communities in dynamic networks, the application of these methods to dynamic heterogeneous networks can be inappropriate, due to the involvement of different types of nodes and edges and the need to treat them differently. In this paper, we propose a statistical framework for detecting common communities in dynamic and heterogeneous networks. Under this framework, we develop a fast community detection method called DHNet that can efficiently estimate the community label as well as the number of communities. An attractive feature of DHNet is that it does not require the number of communities to be known a priori, a common assumption in community detection methods. While DHNet does not require any parametric assumptions on the underlying network model, we show that the identified label is consistent under a time-varying heterogeneous stochastic block model with a temporal correlation structure and edge sparsity. We further illustrate the utility of DHNet through simulations and an application to review data from Yelp, where DHNet shows improvements both in terms of accuracy and interpretability over existing solutions.
翻译:动态的多元网络描述不同类型节点和边缘相互作用的时间演变情况。虽然在动态网络中找到社区方面有大量文献,但将这些方法应用于动态的多元网络可能不适当,因为有不同类型的节点和边缘参与,而且需要区别对待这些网络。在本文件中,我们提议了一个统计框架,用于在动态和多元网络中探测共同社区。在这个框架内,我们开发了一个称为DHNet的快速社区检测方法,可以有效地估计社区标签和社区数量。DHNet的一个有吸引力的特征是,它并不要求事先知道社区的数目,这是社区探测方法的一个共同假设。虽然DHNet并不要求在基本网络模型上作任何参数假设,但我们表明,所找到的标签在一个有时间变化的多变的多变性区块模型下,具有时间相关性结构和边缘宽阔度。我们进一步通过模拟和用于审查Yelp的数据来说明DHNet的效用,在那里,DHNet显示对现有解决办法的准确性和可解释性都有改进。