论文标题
通过图形卷积网络和提取的侧面信息进行半监督节点分类
Semi-Supervised Node Classification by Graph Convolutional Networks and Extracted Side Information
论文作者
论文摘要
与图中的其他节点相比，群集中存在的图的节点更有可能彼此连接。然后揭示有关某些节点的一些信息，图表的结构（图边）提供了这个机会，可以了解有关其他节点的更多信息。从这个角度来看，本文通过图形卷积网络（GCN）在半监督方案中重新审视了节点分类任务。目的是从围绕显示的节点标签循环的信息流中受益。本文的贡献是双重的。首先，本文提供了一种从图形实现中提取侧面信息的方法。然后提出了新的GCN体系结构，结合了传统GCN的输出和提取的侧面信息。本文的另一个贡献与在许多应用程序中图形实现旁边存在的非刻画观察（独立侧面信息）有关。实际上，提取的侧面信息可以用独立于图形结构的侧面信息替换。在这两种情况下，关于合成和现实世界数据集的实验表明，与现有的节点分类任务相比，所提出的模型可实现更高的预测准确性。
The nodes of a graph existing in a cluster are more likely to connect to each other than with other nodes in the graph. Then revealing some information about some nodes, the structure of the graph (graph edges) provides this opportunity to know more information about other nodes. From this perspective, this paper revisits the node classification task in a semi-supervised scenario by graph convolutional networks (GCNs). The goal is to benefit from the flow of information that circulates around the revealed node labels. The contribution of this paper is twofold. First, this paper provides a method for extracting side information from a graph realization. Then a new GCN architecture is presented that combines the output of traditional GCN and the extracted side information. Another contribution of this paper is relevant to non-graph observations (independent side information) that exists beside a graph realization in many applications. Indeed, the extracted side information can be replaced by a sequence of side information that is independent of the graph structure. For both cases, the experiments on synthetic and real-world datasets demonstrate that the proposed model achieves a higher prediction accuracy in comparison to the existing state-of-the-art methods for the node classification task.
