Signed Graph Convolutional Networks

By Tyler Derr, Yao Ma, Jiliang Tang. 2018

325 citations recorded by the source corpus.

Due to the fact much of today's data can be represented as graphs, there has been a demand for generalizing neural network models for graph data. One recent direction that has shown fruitful results, and therefore growing interest, is the usage of graph convolutional neural networks (GCNs). They have been shown to provide a significant improvement on a wide range of tasks in network analysis, one of which being node representation learning. The task of learning low-dimensional node representations has shown to increase performance on a plethora of other tasks from link prediction and node classification, to community detection and visualization. Simultaneously, signed networks (or graphs having both positive and negative links) have become ubiquitous with the growing popularity of social media. However, since previous GCN models have primarily focused on unsigned networks (or graphs consisting of only positive links), it is unclear how they could be applied to signed networks due to the challenges presented by negative links. The primary challenges are based on negative links having not only a different semantic meaning as compared to positive links, but their principles are inherently different and they form complex relations with positive links. Therefore we propose a dedicated and principled effort that utilizes balance theory to correctly aggregate and propagate the information across layers of a signed GCN model. We perform empirical experiments comparing our proposed signed GCN against state-of-the-art baselines for learning node representations in signed networks. More specifically, our experiments are performed on four real-world datasets for the classical link sign prediction problem that is commonly used as the benchmark for signed network embeddings algorithms.

Subfields: Artificial Intelligence, Information Systems, Statistical and Nonlinear Physics, Graph Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

In Statistical and Nonlinear Physics

In Graph Learning

Citations in the corpus

Fast Graph Representation Learning with PyTorch Geometric (2019, background): “, 2019), the Dynamic Neighborhood Aggregation (DNA) operator (Fey, 2019) and the signed operator for learning in signed networks (Derr et al., 2018).”
A Comprehensive Survey on Graph Neural Networks (2019, background): “Spatial-based models are more flexible to handle multi-source graph inputs such as edge inputs [15], [27], [86], [95], [96], directed graphs [25], [72], signed graphs [97], and heterogeneous graphs [98], [99], because these graph inputs can be incorporated into the aggregation function easily.”
Diffusion Improves Graph Learning (2019, background): “"opposites attract", might be negative edge weights [17, 43].”
Graph Convolutional Networks with EigenPooling (2019, unknown): “Graph neural networks have also been extended to different types of graphs [9, 26, 27] and applied to various applications [12, 28, 35, 41, 45, 47].”
DEBAGREEMENT: A comment-reply dataset for (dis)agreement detection in online debates (2021, methodology): “Modeling online discussion forums as graphs of interactions between users enables researchers to: i) translate the (dis)agreement detection task into a sign link prediction one [45], ii) use recent advances in GRL methods [20], and iii) leverage existing signed graph embedding methods [41, 15, 30] t”
Graph Neural Networks: A Review of Methods and Applications (2018, methodology): “Instead of simply treating the negative edges as the absent edges or another type of edges, SGCN (Derr et al., 2018) utilizes balance theory to capture the interactions between positive edges and negative edges.”
Graph Neural Networks for Social Recommendation (2019, background): “Hence, GNNs naturally integrate the node information as well as the topological structure and have been demonstrated to be powerful in representation learning [5, 7, 15].”
Cross-Modal Hierarchical Modelling for Fine-Grained Sketch Based Image Retrieval (2020, methodology): “While EdgePool [12] reduces almost half its nodes per GNN layer by contracting its edges, EigenPool [10] controls pooling ratio based on graph Fourier transform.”
GNNGuard: Defending Graph Neural Networks against Adversarial Attacks (2020, background): “GNNGUARD works with many GNNs, including Graph Convolutional Network (GCN) [3], Graph Attention Network (GAT) [19], Graph Isomorphism Network (GIN) [7], Jumping Knowledge (JK-Net) [20], GraphSAINT [21], GraphSAGE [36], and SignedGCN [42].”
Deep Learning on Graphs: A Survey (2018, unknown): “have unique structures, and they pose additional challenges to the existing methods [158].”