Multi-view network embedding via graph factorization clustering and co-regularized multi-view agreement

Yiwei Sun, Ngot Bui, Tsung Yu Hsieh, Vasant Honavar

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

1 Citation (Scopus)

Abstract

Real-world social networks and digital platforms are comprised of individuals (nodes) that are linked to other individuals or entities through multiple types of relationships (links). Sub-networks of such a network based on each type of link correspond to distinct views of the underlying network. In real-world applications each node is typically linked to only a small subset of other nodes. Hence, practical approaches to problems such as node labeling have to cope with the resulting sparse networks. While low-dimensional network embeddings offer a promising approach to this problem, most of the current network embedding methods focus primarily on single view networks. We introduce a novel multi-view network embedding (MVNE) algorithm for constructing low-dimensional node embeddings from multi-view networks. MVNE adapts and extends an approach to single view node embedding using graph factorization clustering (GFC) to the multi-view setting using an objective function that maximizes the agreement between views based on both the local and global structure of the underlying multi-view graph. Our experiments with several benchmark real-world single view networks show that SVNE yields network embeddings that are competitive with or superior to those produced by the state-of-the-art single view network embedding methods when the embeddings are used for labeling unlabeled nodes in the networks. Our experiments with several multi-view networks show that MVNE substantially outperforms the single view methods on integrated view and the state-of-the-art multi-view methods. We further show that even when the goal is to predict labels of nodes within a single target view, MVNE outperforms its single-view counterpart suggesting that the MVNE is able to extract the information that is useful for labeling nodes in the target view from the all of the views.

Original languageEnglish (US)
Title of host publicationProceedings - 18th IEEE International Conference on Data Mining Workshops, ICDMW 2018
EditorsZhenhui Li, Hanghang Tong, Feida Zhu, Jeffrey Yu
PublisherIEEE Computer Society
Pages1006-1013
Number of pages8
ISBN (Electronic)9781538692882
DOIs
StatePublished - Feb 7 2019
Event18th IEEE International Conference on Data Mining Workshops, ICDMW 2018 - Singapore, Singapore
Duration: Nov 17 2018Nov 20 2018

Publication series

NameIEEE International Conference on Data Mining Workshops, ICDMW
Volume2018-November
ISSN (Print)2375-9232
ISSN (Electronic)2375-9259

Conference

Conference18th IEEE International Conference on Data Mining Workshops, ICDMW 2018
CountrySingapore
CitySingapore
Period11/17/1811/20/18

Fingerprint

Factorization
Labeling
Labels
Experiments

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Software

Cite this

Sun, Y., Bui, N., Hsieh, T. Y., & Honavar, V. (2019). Multi-view network embedding via graph factorization clustering and co-regularized multi-view agreement. In Z. Li, H. Tong, F. Zhu, & J. Yu (Eds.), Proceedings - 18th IEEE International Conference on Data Mining Workshops, ICDMW 2018 (pp. 1006-1013). [8637384] (IEEE International Conference on Data Mining Workshops, ICDMW; Vol. 2018-November). IEEE Computer Society. https://doi.org/10.1109/ICDMW.2018.00145
Sun, Yiwei ; Bui, Ngot ; Hsieh, Tsung Yu ; Honavar, Vasant. / Multi-view network embedding via graph factorization clustering and co-regularized multi-view agreement. Proceedings - 18th IEEE International Conference on Data Mining Workshops, ICDMW 2018. editor / Zhenhui Li ; Hanghang Tong ; Feida Zhu ; Jeffrey Yu. IEEE Computer Society, 2019. pp. 1006-1013 (IEEE International Conference on Data Mining Workshops, ICDMW).
@inproceedings{94443ea068e84dcbbf134b8800533c52,
title = "Multi-view network embedding via graph factorization clustering and co-regularized multi-view agreement",
abstract = "Real-world social networks and digital platforms are comprised of individuals (nodes) that are linked to other individuals or entities through multiple types of relationships (links). Sub-networks of such a network based on each type of link correspond to distinct views of the underlying network. In real-world applications each node is typically linked to only a small subset of other nodes. Hence, practical approaches to problems such as node labeling have to cope with the resulting sparse networks. While low-dimensional network embeddings offer a promising approach to this problem, most of the current network embedding methods focus primarily on single view networks. We introduce a novel multi-view network embedding (MVNE) algorithm for constructing low-dimensional node embeddings from multi-view networks. MVNE adapts and extends an approach to single view node embedding using graph factorization clustering (GFC) to the multi-view setting using an objective function that maximizes the agreement between views based on both the local and global structure of the underlying multi-view graph. Our experiments with several benchmark real-world single view networks show that SVNE yields network embeddings that are competitive with or superior to those produced by the state-of-the-art single view network embedding methods when the embeddings are used for labeling unlabeled nodes in the networks. Our experiments with several multi-view networks show that MVNE substantially outperforms the single view methods on integrated view and the state-of-the-art multi-view methods. We further show that even when the goal is to predict labels of nodes within a single target view, MVNE outperforms its single-view counterpart suggesting that the MVNE is able to extract the information that is useful for labeling nodes in the target view from the all of the views.",
author = "Yiwei Sun and Ngot Bui and Hsieh, {Tsung Yu} and Vasant Honavar",
year = "2019",
month = "2",
day = "7",
doi = "10.1109/ICDMW.2018.00145",
language = "English (US)",
series = "IEEE International Conference on Data Mining Workshops, ICDMW",
publisher = "IEEE Computer Society",
pages = "1006--1013",
editor = "Zhenhui Li and Hanghang Tong and Feida Zhu and Jeffrey Yu",
booktitle = "Proceedings - 18th IEEE International Conference on Data Mining Workshops, ICDMW 2018",
address = "United States",

}

Sun, Y, Bui, N, Hsieh, TY & Honavar, V 2019, Multi-view network embedding via graph factorization clustering and co-regularized multi-view agreement. in Z Li, H Tong, F Zhu & J Yu (eds), Proceedings - 18th IEEE International Conference on Data Mining Workshops, ICDMW 2018., 8637384, IEEE International Conference on Data Mining Workshops, ICDMW, vol. 2018-November, IEEE Computer Society, pp. 1006-1013, 18th IEEE International Conference on Data Mining Workshops, ICDMW 2018, Singapore, Singapore, 11/17/18. https://doi.org/10.1109/ICDMW.2018.00145

Multi-view network embedding via graph factorization clustering and co-regularized multi-view agreement. / Sun, Yiwei; Bui, Ngot; Hsieh, Tsung Yu; Honavar, Vasant.

Proceedings - 18th IEEE International Conference on Data Mining Workshops, ICDMW 2018. ed. / Zhenhui Li; Hanghang Tong; Feida Zhu; Jeffrey Yu. IEEE Computer Society, 2019. p. 1006-1013 8637384 (IEEE International Conference on Data Mining Workshops, ICDMW; Vol. 2018-November).

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

TY - GEN

T1 - Multi-view network embedding via graph factorization clustering and co-regularized multi-view agreement

AU - Sun, Yiwei

AU - Bui, Ngot

AU - Hsieh, Tsung Yu

AU - Honavar, Vasant

PY - 2019/2/7

Y1 - 2019/2/7

N2 - Real-world social networks and digital platforms are comprised of individuals (nodes) that are linked to other individuals or entities through multiple types of relationships (links). Sub-networks of such a network based on each type of link correspond to distinct views of the underlying network. In real-world applications each node is typically linked to only a small subset of other nodes. Hence, practical approaches to problems such as node labeling have to cope with the resulting sparse networks. While low-dimensional network embeddings offer a promising approach to this problem, most of the current network embedding methods focus primarily on single view networks. We introduce a novel multi-view network embedding (MVNE) algorithm for constructing low-dimensional node embeddings from multi-view networks. MVNE adapts and extends an approach to single view node embedding using graph factorization clustering (GFC) to the multi-view setting using an objective function that maximizes the agreement between views based on both the local and global structure of the underlying multi-view graph. Our experiments with several benchmark real-world single view networks show that SVNE yields network embeddings that are competitive with or superior to those produced by the state-of-the-art single view network embedding methods when the embeddings are used for labeling unlabeled nodes in the networks. Our experiments with several multi-view networks show that MVNE substantially outperforms the single view methods on integrated view and the state-of-the-art multi-view methods. We further show that even when the goal is to predict labels of nodes within a single target view, MVNE outperforms its single-view counterpart suggesting that the MVNE is able to extract the information that is useful for labeling nodes in the target view from the all of the views.

AB - Real-world social networks and digital platforms are comprised of individuals (nodes) that are linked to other individuals or entities through multiple types of relationships (links). Sub-networks of such a network based on each type of link correspond to distinct views of the underlying network. In real-world applications each node is typically linked to only a small subset of other nodes. Hence, practical approaches to problems such as node labeling have to cope with the resulting sparse networks. While low-dimensional network embeddings offer a promising approach to this problem, most of the current network embedding methods focus primarily on single view networks. We introduce a novel multi-view network embedding (MVNE) algorithm for constructing low-dimensional node embeddings from multi-view networks. MVNE adapts and extends an approach to single view node embedding using graph factorization clustering (GFC) to the multi-view setting using an objective function that maximizes the agreement between views based on both the local and global structure of the underlying multi-view graph. Our experiments with several benchmark real-world single view networks show that SVNE yields network embeddings that are competitive with or superior to those produced by the state-of-the-art single view network embedding methods when the embeddings are used for labeling unlabeled nodes in the networks. Our experiments with several multi-view networks show that MVNE substantially outperforms the single view methods on integrated view and the state-of-the-art multi-view methods. We further show that even when the goal is to predict labels of nodes within a single target view, MVNE outperforms its single-view counterpart suggesting that the MVNE is able to extract the information that is useful for labeling nodes in the target view from the all of the views.

UR - http://www.scopus.com/inward/record.url?scp=85062818085&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85062818085&partnerID=8YFLogxK

U2 - 10.1109/ICDMW.2018.00145

DO - 10.1109/ICDMW.2018.00145

M3 - Conference contribution

AN - SCOPUS:85062818085

T3 - IEEE International Conference on Data Mining Workshops, ICDMW

SP - 1006

EP - 1013

BT - Proceedings - 18th IEEE International Conference on Data Mining Workshops, ICDMW 2018

A2 - Li, Zhenhui

A2 - Tong, Hanghang

A2 - Zhu, Feida

A2 - Yu, Jeffrey

PB - IEEE Computer Society

ER -

Sun Y, Bui N, Hsieh TY, Honavar V. Multi-view network embedding via graph factorization clustering and co-regularized multi-view agreement. In Li Z, Tong H, Zhu F, Yu J, editors, Proceedings - 18th IEEE International Conference on Data Mining Workshops, ICDMW 2018. IEEE Computer Society. 2019. p. 1006-1013. 8637384. (IEEE International Conference on Data Mining Workshops, ICDMW). https://doi.org/10.1109/ICDMW.2018.00145