Overlapping community detection has become a very hot research topic in recent decades, and a plethora of methods have been proposed. But, a common challenge in many existing overlapping community detection approaches is that the number of communities K must be predefined manually. We propose a flexible nonparametric Bayesian generative model for count-value networks, which can allow K to increase as more and more data are encountered instead of to be fixed in advance. The Indian buffet process was used to model the community assignment matrix Z, and an uncollapsed Gibbs sampler has been derived. However, as the community assignment matrix Z is a structured multi-variable parameter, how to summarize the posterior inference results and estimate the inference quality about Z, is still a considerable challenge in the literature. In this paper, a graph convolutional neural network based graph classifier was utilized to help to summarize the results and to estimate the inference quality about Z. We conduct extensive experiments on synthetic data and real data, and find that empirically, the traditional posterior summarization strategy is reliable.

中文翻译：

---

估计重叠社区相关无穷潜在特征模型的后验推断质量

重叠的社区检测已成为近几十年来非常热门的研究主题，并且已经提出了许多方法。但是，在许多现有的重叠社区检测方法中的共同挑战是必须手动预定义社区K的数量。我们为计数值网络提出了一种灵活的非参数贝叶斯生成模型，该模型可以使K随着遇到越来越多的数据而增加，而不必事先确定。印度自助餐过程用于建模社区分配矩阵Z，并且已导出未折叠的Gibbs采样器。但是，作为社区分配矩阵Z作为一个结构化的多变量参数，如何总结后验结果并估计关于Z的推理质量，在文献中仍然是一个相当大的挑战。本文利用基于图卷积神经网络的图分类器，对结果进行总结和估计Z的推理质量。我们对合成数据和真实数据进行了广泛的实验，从经验上发现，传统的后验汇总策略是可靠的。