Path-enhanced graph convolutional networks for node classification without features

Most current graph neural networks (GNNs) are designed from the view of methodology and rarely consider the inherent characters of graph. Although the inherent characters may impact the performance of GNNs, very few methods are proposed to resolve the issue. In this work, we mainly focus on improvin...

Full description

Saved in:
Bibliographic Details
Published in:PloS one 2023-06, Vol.18 (6), p.e0287001-e0287001
Main Authors: Jiao, Qingju, Zhao, Peige, Zhang, Hanjin, Han, Yahong, Liu, Guoying
Format: Article
Language:English
Subjects:
Analysis
Artificial neural networks
Biology and Life Sciences
Classification
Computational linguistics
Computer and Information Sciences
Deep learning
Diffusion Magnetic Resonance Imaging
Earth Sciences
Ecology and Environmental Sciences
Graph neural networks
Graphs
Identification and classification
Language processing
Lymph nodes
Machine learning
Natural language interfaces
Neural networks
Neural Networks, Computer
Nodes
Performance enhancement
Physical Sciences
Records
Shortest-path problems
Social Sciences
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Most current graph neural networks (GNNs) are designed from the view of methodology and rarely consider the inherent characters of graph. Although the inherent characters may impact the performance of GNNs, very few methods are proposed to resolve the issue. In this work, we mainly focus on improving the performance of graph convolutional networks (GCNs) on the graphs without node features. In order to resolve the issue, we propose a method called t-hopGCN to describe t-hop neighbors by the shortest path between two nodes, then the adjacency matrix of t-hop neighbors as features to perform node classification. Experimental results show that t-hopGCN can significantly improve the performance of node classification in the graphs without node features. More importantly, adding the adjacency matrix of t-hop neighbors can improve the performance of existing popular GNNs on node classification.
ISSN:1932-6203
1932-6203
DOI:10.1371/journal.pone.0287001