Label-Consistency based Graph Neural Networks for Semi-supervised Node Classification

TL;DR

This paper introduces LC-GNN, a novel graph neural network that enhances semi-supervised node classification by utilizing label consistency among unconnected nodes, improving performance especially in sparse label scenarios.

Contribution

The paper proposes LC-GNN, which leverages label consistency among unconnected nodes to expand receptive fields, addressing limitations of traditional GNNs that rely solely on connected nodes.

Findings

LC-GNN outperforms traditional GNNs on benchmark datasets.

LC-GNN is especially effective in sparse label scenarios.

Experimental results demonstrate improved accuracy in semi-supervised node classification.

Abstract

Graph neural networks (GNNs) achieve remarkable success in graph-based semi-supervised node classification, leveraging the information from neighboring nodes to improve the representation learning of target node. The success of GNNs at node classification depends on the assumption that connected nodes tend to have the same label. However, such an assumption does not always work, limiting the performance of GNNs at node classification. In this paper, we propose label-consistency based graph neural network(LC-GNN), leveraging node pairs unconnected but with the same labels to enlarge the receptive field of nodes in GNNs. Experiments on benchmark datasets demonstrate the proposed LC-GNN outperforms traditional GNNs in graph-based semi-supervised node classification.We further show the superiority of LC-GNN in sparse scenarios with only a handful of labeled nodes.