Improving link prediction accuracy of network embedding algorithms via rich node attribute information

TL;DR

This work tackles link prediction by enriching network embeddings with rich node attribute information. It introduces AGEE, a plug‑and‑play framework that builds a feature graph from node attributes using feature information entropy and then jointly trains the feature and structure graphs, blending their predictions via a consensus parameter $α$. Empirical results on three citation networks show that AGEE improves AUC by about 3% over strong baselines and yields substantial gains when training data are limited, with an optimal $α$ around 0.6. The approach demonstrates that explicitly leveraging attribute information through a dedicated feature graph can substantially enhance link prediction without redesigning core embedding architectures, offering practical benefits for biology, neuroscience, and social networks where attributes carry latent relational signals.

Abstract

Complex networks are widely used to represent an abundance of real-world relations ranging from social networks to brain networks. Inferring missing links or predicting future ones based on the currently observed network is known as the link prediction task.Recent network embedding based link prediction algorithms have demonstrated ground-breaking performance on link prediction accuracy. Those algorithms usually apply node attributes as the initial feature input to accelerate the convergence speed during the training process. However, they do not take full advantage of node feature information. In this paper,besides applying feature attributes as the initial input, we make better utilization of node attribute information by building attributable networks and plugging attributable networks into some typical link prediction algorithms and naming this algorithm Attributive Graph Enhanced Embedding (AGEE). AGEE is able to automatically learn the weighting trades-off between the structure and the attributive networks. Numerical experiments show that AGEE can improve the link prediction accuracy by around 3% compared with link prediction framework SEAL, Variational Graph AutoEncoder (VGAE), and Node2vec.

Figures (5)

• Figure 1: The attributes distribution of the Cora (left panel) and CiteSeer (right panel) feature matrix with a large number of attributes occurring occasionally, while a small number of attributes contribute as features of many nodes.
• Figure 2: Comparison of feature-information quantified by eq.\ref{['eq:self-information']} and feature occurrence probability of Cora (left panel) and CiteSeer (right panel) datasets.
• Figure 3: The overall architecture of AGEE algorithm. We use a simple seven-node graph with five nodal features to illustrate AGEE’s architecture. In the first stage, based on the feature matrix we build the feature graph according to Eq. \ref{['eq:self-information']}, \ref{['eq:entropy_matrix']} and \ref{['eq:adj']}, we also build the structure graph according to the adjacency matrix, in the structure graph, we use the dashed lines to indicate some of the representative edges that need to be predicted by link prediction algorithms and apply solid lines to represent edges used to train the link prediction algorithms. In the structure graph, there are no edges between nodes 0 and 6, 5 and 0, in the feature graph, there are links between them. The featured graph works as a supplementary for the structure graph. In the second stage, we train the feature graph and structure graph separately and use a hyper-parameter $\alpha$ to find the trade-off between the structure link prediction probability and the feature link prediction probability.
• Figure 4: AUC score of link prediction accuracy comparison between original node2vec algorithm and AGEE_node2vec on a variety of training sets for Cora(left panel) and CiteSeer (right panel).
• Figure 5: AUC score of link prediction accuracy across different prediction algorithms on Cora and CiteSeer network over hyper-parameter $\alpha$.