A Network Science perspective of Graph Convolutional Networks: A survey

Abstract

The mining and exploitation of graph structural information have been the focal points in the study of complex networks. Traditional structural measures in Network Science focus on the analysis and modelling of complex networks from the perspective of network structure, such as the centrality measures, the clustering coefficient, and motifs and graphlets, and they have become basic tools for studying and understanding graphs. In comparison, graph neural networks, especially graph convolutional networks (GCNs), are particularly effective at integrating node features into graph structures via neighbourhood aggregation and message passing, and have been shown to significantly improve the performances in a variety of learning tasks. These two classes of methods are, however, typically treated separately with limited references to each other. In this work, aiming to establish relationships between them, we provide a network science perspective of GCNs. Our novel taxonomy classifies GCNs from three structural information angles, i.e., the layer-wise message aggregation scope, the message content, and the overall learning scope. Moreover, as a prerequisite for reviewing GCNs via a network science perspective, we also summarise traditional structural measures and propose a new taxonomy for them. Finally and most importantly, we draw connections between traditional structural approaches and graph convolutional networks, and discuss potential directions for future research.

References in corpus (36)

• Fast unfolding of communities in large networks
• The structure and dynamics of multilayer networks
• Simplifying Graph Convolutional Networks
• Vital nodes identification in complex networks
• Leaders in Social Networks, the Delicious Case
• Clustering in Complex Directed Networks
• New Model of Internet Topology Using k-shell Decomposition
• Effective and Efficient Similarity Index for Link Prediction of Complex Networks
• Complex networks in climate dynamics - Comparing linear and nonlinear network construction methods
• Graph Self-Supervised Learning: A Survey
• The Slashdot Zoo: Mining a Social Network with Negative Edges
• MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixing
• Spreading processes in Multilayer Networks
• The clustering coefficient and community structure of bipartite networks
• A Novel Measure of Edge Centrality in Social Networks
• A Survey of Natural Language Generation
• Communicability Betweenness in Complex Networks
• Lobby index in networks
• Self-supervised Learning on Graphs: Deep Insights and New Direction
• Learning to Generate Questions by Learning What not to Generate
• Distance Encoding: Design Provably More Powerful Neural Networks for Graph Representation Learning
• A degree centrality in multi-layered social network
• Directed Graph Convolutional Network
• DIG: A Turnkey Library for Diving into Graph Deep Learning Research
• Decoupling the Depth and Scope of Graph Neural Networks
• Neo-GNNs: Neighborhood Overlap-aware Graph Neural Networks for Link Prediction
• Nested Graph Neural Networks
• Edge-Featured Graph Attention Network
• From Stars to Subgraphs: Uplifting Any GNN with Local Structure Awareness
• Graph Neural Networks with Local Graph Parameters
• TF-GNN: Graph Neural Networks in TensorFlow
• Clustering as a measure of the local topology of networks
• Heterogeneous Graph Matching Networks
• CogDL: A Comprehensive Library for Graph Deep Learning
• Automorphic Equivalence-aware Graph Neural Network
• Asymmetric Graph Representation Learning