Abstract

Graph-structured data naturally appear in numerous application domains, ranging from social analysis, bioinformatics to computer vision. The unique capability of graphs enables capturing the structural relations among data, and thus allows to harvest more insights compared to analyzing data in isolation. However, graph mining is a challenging task due to the underlying complex and diverse connectivity patterns. A potential solution is to learn the representation of a graph in a low-dimensional Euclidean space via embedding techniques that preserve the graph properties. Although tremendous efforts have been made to address the graph representation learning problem, many of them still suffer from their shallow learning mechanisms. On the other hand, deep learning models on graphs have recently emerged in both machine learning and data mining areas and demonstrated superior performance for various problems. In this survey, we conduct a comprehensive review specifically on the emerging field of graph convolutional networks, which is one of the most prominent graph deep learning models. We first introduce two taxonomies to group the existing works based on the types of convolutions and the areas of applications, then highlight some graph convolutional network models in details. Finally, we present several challenges in this area and discuss potential directions for future research.

Original languageEnglish (US)
Title of host publicationComputational Data and Social Networks - 7th International Conference, CSoNet 2018, Proceedings
EditorsMy T. Thai, Xuemin Chen, Wei Wayne Li, Arunabha Sen
PublisherSpringer Verlag
Pages79-91
Number of pages13
ISBN (Print)9783030046477
DOIs
StatePublished - Jan 1 2018
Event7th International Conference on Computational Data and Social Networks, CSoNet 2018 - Shanghai, China
Duration: Dec 18 2018Dec 20 2018

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11280 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other7th International Conference on Computational Data and Social Networks, CSoNet 2018
CountryChina
CityShanghai
Period12/18/1812/20/18

Fingerprint

Network Algorithms
Graph in graph theory
Taxonomies
Bioinformatics
Convolution
Computer vision
Data mining
Learning systems
Graph Mining
Graph Representation
Taxonomy
Computer Vision
Network Model
Isolation
Euclidean space
Data Mining
Machine Learning
Connectivity
Deep learning
Learning

Keywords

  • Graph convolutional networks
  • Spatial
  • Spectral

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Zhang, S., Tong, H., Xu, J., & Maciejewski, R. (2018). Graph convolutional networks: algorithms, applications and open challenges. In M. T. Thai, X. Chen, W. W. Li, & A. Sen (Eds.), Computational Data and Social Networks - 7th International Conference, CSoNet 2018, Proceedings (pp. 79-91). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11280 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-030-04648-4_7

Graph convolutional networks : algorithms, applications and open challenges. / Zhang, Si; Tong, Hanghang; Xu, Jiejun; Maciejewski, Ross.

Computational Data and Social Networks - 7th International Conference, CSoNet 2018, Proceedings. ed. / My T. Thai; Xuemin Chen; Wei Wayne Li; Arunabha Sen. Springer Verlag, 2018. p. 79-91 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11280 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zhang, S, Tong, H, Xu, J & Maciejewski, R 2018, Graph convolutional networks: algorithms, applications and open challenges. in MT Thai, X Chen, WW Li & A Sen (eds), Computational Data and Social Networks - 7th International Conference, CSoNet 2018, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11280 LNCS, Springer Verlag, pp. 79-91, 7th International Conference on Computational Data and Social Networks, CSoNet 2018, Shanghai, China, 12/18/18. https://doi.org/10.1007/978-3-030-04648-4_7
Zhang S, Tong H, Xu J, Maciejewski R. Graph convolutional networks: algorithms, applications and open challenges. In Thai MT, Chen X, Li WW, Sen A, editors, Computational Data and Social Networks - 7th International Conference, CSoNet 2018, Proceedings. Springer Verlag. 2018. p. 79-91. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-030-04648-4_7
Zhang, Si ; Tong, Hanghang ; Xu, Jiejun ; Maciejewski, Ross. / Graph convolutional networks : algorithms, applications and open challenges. Computational Data and Social Networks - 7th International Conference, CSoNet 2018, Proceedings. editor / My T. Thai ; Xuemin Chen ; Wei Wayne Li ; Arunabha Sen. Springer Verlag, 2018. pp. 79-91 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{b63bb65902f943a5b796abcc7b52d3e0,
title = "Graph convolutional networks: algorithms, applications and open challenges",
abstract = "Graph-structured data naturally appear in numerous application domains, ranging from social analysis, bioinformatics to computer vision. The unique capability of graphs enables capturing the structural relations among data, and thus allows to harvest more insights compared to analyzing data in isolation. However, graph mining is a challenging task due to the underlying complex and diverse connectivity patterns. A potential solution is to learn the representation of a graph in a low-dimensional Euclidean space via embedding techniques that preserve the graph properties. Although tremendous efforts have been made to address the graph representation learning problem, many of them still suffer from their shallow learning mechanisms. On the other hand, deep learning models on graphs have recently emerged in both machine learning and data mining areas and demonstrated superior performance for various problems. In this survey, we conduct a comprehensive review specifically on the emerging field of graph convolutional networks, which is one of the most prominent graph deep learning models. We first introduce two taxonomies to group the existing works based on the types of convolutions and the areas of applications, then highlight some graph convolutional network models in details. Finally, we present several challenges in this area and discuss potential directions for future research.",
keywords = "Graph convolutional networks, Spatial, Spectral",
author = "Si Zhang and Hanghang Tong and Jiejun Xu and Ross Maciejewski",
year = "2018",
month = "1",
day = "1",
doi = "10.1007/978-3-030-04648-4_7",
language = "English (US)",
isbn = "9783030046477",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "79--91",
editor = "Thai, {My T.} and Xuemin Chen and Li, {Wei Wayne} and Arunabha Sen",
booktitle = "Computational Data and Social Networks - 7th International Conference, CSoNet 2018, Proceedings",

}

TY - GEN

T1 - Graph convolutional networks

T2 - algorithms, applications and open challenges

AU - Zhang, Si

AU - Tong, Hanghang

AU - Xu, Jiejun

AU - Maciejewski, Ross

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Graph-structured data naturally appear in numerous application domains, ranging from social analysis, bioinformatics to computer vision. The unique capability of graphs enables capturing the structural relations among data, and thus allows to harvest more insights compared to analyzing data in isolation. However, graph mining is a challenging task due to the underlying complex and diverse connectivity patterns. A potential solution is to learn the representation of a graph in a low-dimensional Euclidean space via embedding techniques that preserve the graph properties. Although tremendous efforts have been made to address the graph representation learning problem, many of them still suffer from their shallow learning mechanisms. On the other hand, deep learning models on graphs have recently emerged in both machine learning and data mining areas and demonstrated superior performance for various problems. In this survey, we conduct a comprehensive review specifically on the emerging field of graph convolutional networks, which is one of the most prominent graph deep learning models. We first introduce two taxonomies to group the existing works based on the types of convolutions and the areas of applications, then highlight some graph convolutional network models in details. Finally, we present several challenges in this area and discuss potential directions for future research.

AB - Graph-structured data naturally appear in numerous application domains, ranging from social analysis, bioinformatics to computer vision. The unique capability of graphs enables capturing the structural relations among data, and thus allows to harvest more insights compared to analyzing data in isolation. However, graph mining is a challenging task due to the underlying complex and diverse connectivity patterns. A potential solution is to learn the representation of a graph in a low-dimensional Euclidean space via embedding techniques that preserve the graph properties. Although tremendous efforts have been made to address the graph representation learning problem, many of them still suffer from their shallow learning mechanisms. On the other hand, deep learning models on graphs have recently emerged in both machine learning and data mining areas and demonstrated superior performance for various problems. In this survey, we conduct a comprehensive review specifically on the emerging field of graph convolutional networks, which is one of the most prominent graph deep learning models. We first introduce two taxonomies to group the existing works based on the types of convolutions and the areas of applications, then highlight some graph convolutional network models in details. Finally, we present several challenges in this area and discuss potential directions for future research.

KW - Graph convolutional networks

KW - Spatial

KW - Spectral

UR - http://www.scopus.com/inward/record.url?scp=85059053261&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85059053261&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-04648-4_7

DO - 10.1007/978-3-030-04648-4_7

M3 - Conference contribution

SN - 9783030046477

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 79

EP - 91

BT - Computational Data and Social Networks - 7th International Conference, CSoNet 2018, Proceedings

A2 - Thai, My T.

A2 - Chen, Xuemin

A2 - Li, Wei Wayne

A2 - Sen, Arunabha

PB - Springer Verlag

ER -