Negative Samples-enhanced Graph Convolutional Neural Networks

Duan, W; Xuan, J; Lu, J

Negative Samples-enhanced Graph Convolutional Neural Networks

Duan, W

Xuan, J

Lu, J

Permalink

Publisher:: IEEE
Publication Type:: Conference Proceeding
Citation:: 2021 IEEE International Conference on Intelligent Systems and Knowledge Engineering, ISKE 2021, 2022, 00, pp. 262-268
Issue Date:: 2022-01-01

Closed Access

	Filename	Description	Size
	Negative_Samples-enhanced_Graph_Convolutional_Neural_Networks.pdf	Published version	2.4 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Duan, W https://orcid.org/0000-0002-2185-0333
dc.contributor.author	Xuan, J https://orcid.org/0000-0002-8367-6908
dc.contributor.author	Lu, J https://orcid.org/0000-0003-0690-4732
dc.date	2021-11-26
dc.date.accessioned	2023-03-23T05:51:34Z
dc.date.available	2023-03-23T05:51:34Z
dc.date.issued	2022-01-01
dc.identifier.citation	2021 IEEE International Conference on Intelligent Systems and Knowledge Engineering, ISKE 2021, 2022, 00, pp. 262-268
dc.identifier.isbn	9781665405539
dc.identifier.uri	http://hdl.handle.net/10453/168227
dc.description.abstract	Graph neural networks (GNNs) have shown great success in graph representation learning by extracting high-level features from nodes and their topology. Many previous studies have used the message passing mechanism to fuse neighborhood information with an assumption that neighbors should share some similar behaviors and then have positive correlations. Therefore, the neighbor nodes can be considered as positive samples and most of existing methods can be viewed as using these positive samples to update new feature vectors. Along with these positive samples, there are also some hidden negative samples (non-neighborhood nodes) which also contains information for a certain node but has been ignored so far. In addition, existing studies on negative sampling for graphs only fuse the sampling results into the loss function for the model training. In this paper, we propose a Negative Samples-enhanced Graph Convolutional Neural Networks (NegGCNs), where the negatively sampled nodes are directly incorporated into the message passing mechanism and used to update new node feature vectors. We further investigated the effects of negative sampling rate and the number of negatively sampled nodes on the classification accuracy of graph nodes. Experiments on citation network datasets suggest that the proposed NegGCNs model can improve the accuracy of graph node classification with a specific negative rate compared to GCNs.
dc.language	en
dc.publisher	IEEE
dc.relation	http://purl.org/au-research/grants/arc/FL190100149
dc.relation	http://purl.org/au-research/grants/arc/DE200100245
dc.relation.ispartof	2021 IEEE International Conference on Intelligent Systems and Knowledge Engineering, ISKE 2021
dc.relation.ispartof	International Conference on Intelligent Systems and Knowledge Engineering
dc.relation.isbasedon	10.1109/ISKE54062.2021.9755440
dc.rights	info:eu-repo/semantics/closedAccess
dc.title	Negative Samples-enhanced Graph Convolutional Neural Networks
dc.type	Conference Proceeding
utslib.citation.volume	00
utslib.location.activity	Chengdu, China
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Strength - AAII - Australian Artificial Intelligence Institute
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Computer Science
utslib.copyright.status	closed_access	*
pubs.consider-herdc	false
dc.date.updated	2023-03-23T05:51:33Z
pubs.finish-date	2021-11-28
pubs.place-of-publication	Piscataway, USA
pubs.publication-status	Published
pubs.start-date	2021-11-26
pubs.volume	00
dc.location	Piscataway, USA

Abstract:

Graph neural networks (GNNs) have shown great success in graph representation learning by extracting high-level features from nodes and their topology. Many previous studies have used the message passing mechanism to fuse neighborhood information with an assumption that neighbors should share some similar behaviors and then have positive correlations. Therefore, the neighbor nodes can be considered as positive samples and most of existing methods can be viewed as using these positive samples to update new feature vectors. Along with these positive samples, there are also some hidden negative samples (non-neighborhood nodes) which also contains information for a certain node but has been ignored so far. In addition, existing studies on negative sampling for graphs only fuse the sampling results into the loss function for the model training. In this paper, we propose a Negative Samples-enhanced Graph Convolutional Neural Networks (NegGCNs), where the negatively sampled nodes are directly incorporated into the message passing mechanism and used to update new node feature vectors. We further investigated the effects of negative sampling rate and the number of negatively sampled nodes on the classification accuracy of graph nodes. Experiments on citation network datasets suggest that the proposed NegGCNs model can improve the accuracy of graph node classification with a specific negative rate compared to GCNs.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/168227