Abstract:Signed networks are mathematical structures that encode positive and negative relations between entities such as friend/foe or trust/distrust. Recently, several papers studied the construction of useful low-dimensional representations (embeddings) of these networks for the prediction of missing relations or signs. Existing embedding methods for sign prediction generally enforce different notions of status or balance theories in their optimization function. These theories, however, are often inaccurate or incomplete, which negatively impacts method performance. In this context, we introduce conditional signed network embedding (CSNE). Our probabilistic approach models structural information about the signs in the network separately from fine-grained detail. Structural information is represented in the form of a prior, while the embedding itself is used for capturing fine-grained information. These components are then integrated in a rigorous manner. CSNE's accuracy depends on the existence of sufficiently powerful structural priors for modelling signed networks, currently unavailable in the literature. Thus, as a second main contribution, which we find to be highly valuable in its own right, we also introduce a novel approach to construct priors based on the Maximum Entropy (MaxEnt) principle. These priors can model the \emph{polarity} of nodes (degree to which their links are positive) as well as signed \emph{triangle counts} (a measure of the degree structural balance holds to in a network). Experiments on a variety of real-world networks confirm that CSNE outperforms the state-of-the-art on the task of sign prediction. Moreover, the MaxEnt priors on their own, while less accurate than full CSNE, achieve accuracies competitive with the state-of-the-art at very limited computational cost, thus providing an excellent runtime-accuracy trade-off in resource-constrained situations.

An unclosed structures-preserving embedding model for signed networks

Deep Network Embedding for Graph Representation Learning in Signed Networks

Structure information learning for neutral links in signed network embedding

Signed Network Embedding with Dynamic Metric Learning.

Link Prediction with Signed Latent Factors in Signed Social Networks

Learning Signed Network Embedding Via Graph Attention.

Interpretable Signed Link Prediction with Signed Infomax Hyperbolic Graph

Signed Network Embedding with Application to Simultaneous Detection of Communities and Anomalies

CSNE: Conditional Signed Network Embedding

Solving Link-Oriented Tasks in Signed Network Via an Embedding Approach

Structure-Enhanced Graph Representation Learning for Link Prediction in Signed Networks

SNEGAN: Signed Network Embedding by Using Generative Adversarial Nets

Effective Model Integration Algorithm for Improving Link and Sign Prediction in Complex Networks

Link Prediction in Signed Networks Based on Connection Degree

Status-Aware Signed Heterogeneous Network Embedding With Graph Neural Networks

Learning Node and Edge Embeddings for Signed Networks

Signed Bipartite Graph Neural Networks

Hyperbolic Node Embedding for Signed Networks

A Signed Subgraph Encoding Approach via Linear Optimization for Link Sign Prediction

Enhanced Signed Graph Neural Network with Node Polarity

Network Learning with Directional Sign Patterns