Abstract:Graph Neural Networks are effective in learning representations of graph-structured data. Some recent works are devoted to addressing heterophily, which exists ubiquitously in real-world networks, breaking the homophily assumption that nodes belonging to the same class are more likely to be connected and restricting the generalization of traditional methods in tasks such as node classification. However, these heterophily-oriented methods still lose efficacy in some typical heterophilic datasets. Moreover, issues on leveraging the knowledge from both node features and graph structure and investigating inherent properties of the datasets still need further consideration. In this work, we first provide insights based on similarity metrics to interpret the long-existing confusion that simple models sometimes perform better than models dedicated to heterophilic networks. Then, sticking to these insights and the classification principle of narrowing the intra-class distance and enlarging the inter-class distance of the sample's embeddings, we propose a Similarity-Navigated Graph Neural Network (SNGNN) which uses Node Similarity matrix coupled with mean aggregation operation instead of the normalized adjacency matrix in the neighborhood aggregation process. Moreover, based on SNGNN, a novel explicitly aggregating mechanism for selecting similar neighbors, named SNGNN+, is devised to preserve distinguishable features and handle the heterophilic problem. Additionally, a variant, SNGNN++, is further designed to adaptively integrate the knowledge from both node features and graph structure for improvement. Extensive experiments are conducted and demonstrate that our proposed framework outperforms the state-of-the-art methods for both small-scale and large-scale graphs regardless of their heterophilic extent. Our implementation is available online.

Revisiting Graph Neural Networks for Node Classification in Heterogeneous Graphs.

A Graph Neural Network with Type-Feature Attention for Node Classification on Heterogeneous Graphs

Heterogeneous Graph Tree Networks

Revisiting Heterophily For Graph Neural Networks

Node classification oriented Adaptive Multichannel Heterogeneous Graph Neural Network

Revisiting Attention-Based Graph Neural Networks for Graph Classification

Heterogeneous graph neural networks analysis: a survey of techniques, evaluations and applications

Similarity-navigated Graph Neural Networks for Node Classification

Self-attention Dual Embedding for Graphs with Heterophily

A General Heterogeneous Hypergraph Neural Network for Node Classification

Addressing Heterogeneity and Heterophily in Graphs: A Heterogeneous Heterophilic Spectral Graph Neural Network

Seeking Similarities While Removing Differences: Graph Neural Networks Based on Node Correlation.

When Do Graph Neural Networks Help with Node Classification? Investigating the Impact of Homophily Principle on Node Distinguishability

Disambiguated Node Classification with Graph Neural Networks

Is Heterophily A Real Nightmare For Graph Neural Networks To Do Node Classification?

When Do Graph Neural Networks Help with Node Classification: Investigating the Homophily Principle on Node Distinguishability

SHGNN: Structure-Aware Heterogeneous Graph Neural Network

A gated graph attention network based on dual graph convolution for node embedding

Graph Decoupling Attention Markov Networks for Semisupervised Graph Node Classification

Heterogeneous Graph Propagation Network

Breaking the Entanglement of Homophily and Heterophily in Semi-supervised Node Classification