From complex network to skeleton: M                         j                          -Modified topology potential for node importance identification

Hanning Yuan; Kanokwan Malang; Yuanyuan Lv; Aniwat Phaphuangwittayakul

doi:10.1007/978-3-030-05090-0_35

From complex network to skeleton: M _j -Modified topology potential for node importance identification

Hanning Yuan, Kanokwan Malang^*, Yuanyuan Lv, Aniwat Phaphuangwittayakul

^*Corresponding author for this work

School of Computer Science and Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Citations (Scopus)

Abstract

Node importance identification is a crucial content in studying the substantial information and the inherent behaviors of complex network. On the basis of topological characteristics of nodes in complex network, we introduce the idea of topology potential from data field theory to capture the important nodes and view it as the skeleton nodes. Inspired by an assumption that different mass of node (m _j parameter) reflects different quality and interaction reliability over the network space. We propose TP-KS method that is an improved topology potential algorithm whose m _j is identified by k-shell centrality. The important nodes identified by TP-KS is ranked and verified by SIR epidemic spreading model. Through the theoretical and experimental analysis, it is proved that TP-KS can effectively extract the importance of nodes in complex network. The better results from TP-KS are also confirmed in both real-world networks and artificial random scale-free networks.

Original language	English
Title of host publication	Advanced Data Mining and Applications - 14th International Conference, ADMA 2018, Proceedings
Editors	Guojun Gan, Xue Li, Shuliang Wang, Bohan Li
Publisher	Springer Verlag
Pages	413-427
Number of pages	15
ISBN (Print)	9783030050894
DOIs	https://doi.org/10.1007/978-3-030-05090-0_35
Publication status	Published - 2018
Event	14th International Conference on Advanced Data Mining and Applications, ADMA 2018 - Nanjing, China Duration: 16 Nov 2018 → 18 Nov 2018

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	11323 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	14th International Conference on Advanced Data Mining and Applications, ADMA 2018
Country/Territory	China
City	Nanjing
Period	16/11/18 → 18/11/18

Keywords

Complex network
Node importance evaluation
Skeleton network
Topology potential

Access to Document

10.1007/978-3-030-05090-0_35

Cite this

Yuan, H., Malang, K., Lv, Y., & Phaphuangwittayakul, A. (2018). From complex network to skeleton: M _j -Modified topology potential for node importance identification. In G. Gan, X. Li, S. Wang, & B. Li (Eds.), Advanced Data Mining and Applications - 14th International Conference, ADMA 2018, Proceedings (pp. 413-427). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11323 LNAI). Springer Verlag. https://doi.org/10.1007/978-3-030-05090-0_35

Yuan, Hanning ; Malang, Kanokwan ; Lv, Yuanyuan et al. / From complex network to skeleton : M _j -Modified topology potential for node importance identification. Advanced Data Mining and Applications - 14th International Conference, ADMA 2018, Proceedings. editor / Guojun Gan ; Xue Li ; Shuliang Wang ; Bohan Li. Springer Verlag, 2018. pp. 413-427 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{6cb26254b6e944048669db19d9ae506a,

title = "From complex network to skeleton: M j -Modified topology potential for node importance identification",

abstract = " Node importance identification is a crucial content in studying the substantial information and the inherent behaviors of complex network. On the basis of topological characteristics of nodes in complex network, we introduce the idea of topology potential from data field theory to capture the important nodes and view it as the skeleton nodes. Inspired by an assumption that different mass of node (m j parameter) reflects different quality and interaction reliability over the network space. We propose TP-KS method that is an improved topology potential algorithm whose m j is identified by k-shell centrality. The important nodes identified by TP-KS is ranked and verified by SIR epidemic spreading model. Through the theoretical and experimental analysis, it is proved that TP-KS can effectively extract the importance of nodes in complex network. The better results from TP-KS are also confirmed in both real-world networks and artificial random scale-free networks.",

keywords = "Complex network, Node importance evaluation, Skeleton network, Topology potential",

author = "Hanning Yuan and Kanokwan Malang and Yuanyuan Lv and Aniwat Phaphuangwittayakul",

note = "Publisher Copyright: {\textcopyright} 2018, Springer Nature Switzerland AG.; 14th International Conference on Advanced Data Mining and Applications, ADMA 2018 ; Conference date: 16-11-2018 Through 18-11-2018",

year = "2018",

doi = "10.1007/978-3-030-05090-0_35",

language = "English",

isbn = "9783030050894",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "413--427",

editor = "Guojun Gan and Xue Li and Shuliang Wang and Bohan Li",

booktitle = "Advanced Data Mining and Applications - 14th International Conference, ADMA 2018, Proceedings",

address = "Germany",

}

Yuan, H, Malang, K, Lv, Y & Phaphuangwittayakul, A 2018, From complex network to skeleton: M _j -Modified topology potential for node importance identification. in G Gan, X Li, S Wang & B Li (eds), Advanced Data Mining and Applications - 14th International Conference, ADMA 2018, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11323 LNAI, Springer Verlag, pp. 413-427, 14th International Conference on Advanced Data Mining and Applications, ADMA 2018, Nanjing, China, 16/11/18. https://doi.org/10.1007/978-3-030-05090-0_35

From complex network to skeleton: M _j -Modified topology potential for node importance identification. / Yuan, Hanning; Malang, Kanokwan; Lv, Yuanyuan et al.
Advanced Data Mining and Applications - 14th International Conference, ADMA 2018, Proceedings. ed. / Guojun Gan; Xue Li; Shuliang Wang; Bohan Li. Springer Verlag, 2018. p. 413-427 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11323 LNAI).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - From complex network to skeleton

T2 - 14th International Conference on Advanced Data Mining and Applications, ADMA 2018

AU - Yuan, Hanning

AU - Malang, Kanokwan

AU - Lv, Yuanyuan

AU - Phaphuangwittayakul, Aniwat

PY - 2018

Y1 - 2018

N2 - Node importance identification is a crucial content in studying the substantial information and the inherent behaviors of complex network. On the basis of topological characteristics of nodes in complex network, we introduce the idea of topology potential from data field theory to capture the important nodes and view it as the skeleton nodes. Inspired by an assumption that different mass of node (m j parameter) reflects different quality and interaction reliability over the network space. We propose TP-KS method that is an improved topology potential algorithm whose m j is identified by k-shell centrality. The important nodes identified by TP-KS is ranked and verified by SIR epidemic spreading model. Through the theoretical and experimental analysis, it is proved that TP-KS can effectively extract the importance of nodes in complex network. The better results from TP-KS are also confirmed in both real-world networks and artificial random scale-free networks.

AB - Node importance identification is a crucial content in studying the substantial information and the inherent behaviors of complex network. On the basis of topological characteristics of nodes in complex network, we introduce the idea of topology potential from data field theory to capture the important nodes and view it as the skeleton nodes. Inspired by an assumption that different mass of node (m j parameter) reflects different quality and interaction reliability over the network space. We propose TP-KS method that is an improved topology potential algorithm whose m j is identified by k-shell centrality. The important nodes identified by TP-KS is ranked and verified by SIR epidemic spreading model. Through the theoretical and experimental analysis, it is proved that TP-KS can effectively extract the importance of nodes in complex network. The better results from TP-KS are also confirmed in both real-world networks and artificial random scale-free networks.

KW - Complex network

KW - Node importance evaluation

KW - Skeleton network

KW - Topology potential

UR - http://www.scopus.com/inward/record.url?scp=85059764573&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-05090-0_35

DO - 10.1007/978-3-030-05090-0_35

M3 - Conference contribution

AN - SCOPUS:85059764573

SN - 9783030050894

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

SP - 413

EP - 427

BT - Advanced Data Mining and Applications - 14th International Conference, ADMA 2018, Proceedings

A2 - Gan, Guojun

A2 - Li, Xue

A2 - Wang, Shuliang

A2 - Li, Bohan

PB - Springer Verlag

Y2 - 16 November 2018 through 18 November 2018

ER -

Yuan H, Malang K, Lv Y, Phaphuangwittayakul A. From complex network to skeleton: M _j -Modified topology potential for node importance identification. In Gan G, Li X, Wang S, Li B, editors, Advanced Data Mining and Applications - 14th International Conference, ADMA 2018, Proceedings. Springer Verlag. 2018. p. 413-427. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-05090-0_35