When Noisy Labels Meet Class Imbalance on Graphs: A Graph Augmentation Method with LLM and Pseudo Label

TL;DR

This paper introduces GraphALP, a novel graph augmentation framework utilizing large language models and pseudo-labeling to improve node classification on noisy, class-imbalanced graphs, demonstrating superior performance over existing methods.

Contribution

The paper proposes GraphALP, a new graph augmentation method that leverages LLMs and pseudo-labeling to handle noisy labels and class imbalance in graph node classification.

Findings

GraphALP outperforms state-of-the-art methods on noisy, imbalanced graph datasets.

The LLM-based oversampling effectively generates synthetic minority nodes.

Pseudo-labeling with dynamic weighting reduces label noise impact.

Abstract

Class-imbalanced graph node classification is a practical yet underexplored research problem. Although recent studies have attempted to address this issue, they typically assume clean and reliable labels when processing class-imbalanced graphs. This assumption often violates the nature of real-world graphs, where labels frequently contain noise. Given this gap, this paper systematically investigates robust node classification for class-imbalanced graphs with noisy labels. We propose GraphALP, a novel Graph Augmentation framework based on Large language models (LLMs) and Pseudo-labeling techniques. Specifically, we design an LLM-based oversampling method to generate synthetic minority nodes, producing label-accurate minority nodes to alleviate class imbalance. Based on the class-balanced graphs, we develop a dynamically weighted pseudo-labeling method to obtain high-confidence pseudo labels to reduce label noise ratio. Additionally, we implement a secondary LLM-guided oversampling mechanism to mitigate potential class distribution skew caused by pseudo labels. Experimental results show that GraphALP achieves superior performance over state-of-the-art methods on class-imbalanced graphs with noisy labels.