Finding Heterophilic Neighbors via Confidence-based Subgraph Matching for Semi-supervised Node Classification

TL;DR

This paper introduces a confidence-based subgraph matching approach to identify heterophilic neighbors, enhancing GNN performance in heterophilic graphs by improving label propagation and reducing over-smoothing.

Contribution

The paper presents a novel two-phased algorithm combining subgraph matching and modified label propagation to better handle heterophilic graph structures in GNNs.

Findings

Improves GNN accuracy on heterophilic datasets

Reduces over-smoothing in graph neural networks

Effectively identifies task-irrelevant edges

Abstract

Graph Neural Networks (GNNs) have proven to be powerful in many graph-based applications. However, they fail to generalize well under heterophilic setups, where neighbor nodes have different labels. To address this challenge, we employ a confidence ratio as a hyper-parameter, assuming that some of the edges are disassortative (heterophilic). Here, we propose a two-phased algorithm. Firstly, we determine edge coefficients through subgraph matching using a supplementary module. Then, we apply GNNs with a modified label propagation mechanism to utilize the edge coefficients effectively. Specifically, our supplementary module identifies a certain proportion of task-irrelevant edges based on a given confidence ratio. Using the remaining edges, we employ the widely used optimal transport to measure the similarity between two nodes with their subgraphs. Finally, using the coefficients as supplementary information on GNNs, we improve the label propagation mechanism which can prevent two nodes with smaller weights from being closer. The experiments on benchmark datasets show that our model alleviates over-smoothing and improves performance.