Tackling the Imbalance for GNNs

TL;DR

This paper investigates the impact of class imbalance on GNN performance, introduces a label difference index to quantify imbalance effects, and proposes new loss functions and methods that improve classification accuracy on imbalanced datasets.

Contribution

The study defines the label difference index (LDI) to analyze imbalance effects and proposes four new methods based on LDI to enhance GNN classification performance.

Findings

Three of the proposed methods outperform existing approaches in accuracy.

The LDI effectively quantifies imbalance-related misclassification risk.

The methods are applicable to various GNN architectures and settings.

Abstract

Different from deep neural networks for non-graph data classification, graph neural networks (GNNs) leverage the information exchange between nodes (or samples) when representing nodes. The category distribution shows an imbalance or even a highly-skewed trend on nearly all existing benchmark GNN data sets. The imbalanced distribution will cause misclassification of nodes in the minority classes, and even cause the classification performance on the entire data set to decrease. This study explores the effects of the imbalance problem on the performances of GNNs and proposes new methodologies to solve it. First, a node-level index, namely, the label difference index ($LDI$), is defined to quantitatively analyze the relationship between imbalance and misclassification. The less samples in a class, the higher the value of its average $LDI$; the higher the $LDI$ of a sample, the more likely the sample will be misclassified. We define a new loss and propose four new methods based on $LDI$. Experimental results indicate that the classification accuracies of the three among our proposed four new methods are better in both transductive and inductive settings. The $LDI$ can be applied to other GNNs.