Directed Graph Convolutional Network

TL;DR

This paper introduces DGCN, a novel directed graph convolutional network that extends spectral-based GCNs to directed graphs using proximity measures, improving representation learning and performance on real datasets.

Contribution

The paper proposes a new spectral-based GCN model for directed graphs, incorporating first- and second-order proximity to enhance graph representation learning.

Findings

DGCN outperforms existing methods on citation and co-purchase datasets.

Using both first- and second-order proximity improves encoding of graph information.

DGCN achieves better generalization and accuracy in directed graph tasks.

Abstract

Graph Convolutional Networks (GCNs) have been widely used due to their outstanding performance in processing graph-structured data. However, the undirected graphs limit their application scope. In this paper, we extend spectral-based graph convolution to directed graphs by using first- and second-order proximity, which can not only retain the connection properties of the directed graph, but also expand the receptive field of the convolution operation. A new GCN model, called DGCN, is then designed to learn representations on the directed graph, leveraging both the first- and second-order proximity information. We empirically show the fact that GCNs working only with DGCNs can encode more useful information from graph and help achieve better performance when generalized to other models. Moreover, extensive experiments on citation networks and co-purchase datasets demonstrate the superiority of our model against the state-of-the-art methods.