Understanding stability of noisy networks through centrality measures and local connections

Vladimir Ufimtsev, Soumya Sarkar, Animesh Mukherjee, Sanjukta Bhowmick

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

2 Citations (Scopus)

Abstract

Networks created from real-world data contain some inaccuracies or noise, manifested as small changes in the network structure. An important question is whether these small changes can significantly affect the analysis results. In this paper, we study the effect of noise in changing ranks of the high centrality vertices. We compare, using the Jaccard Index (JI), how many of the top-k high centrality nodes from the original network are also part of the top-k ranked nodes from the noisy network. We deem a network as stable if the JI value is high. We observe two features that affect the stability. First, the stability is dependent on the number of top-ranked vertices considered. When the vertices are ordered according to their centrality values, they group into clusters. Perturbations to the network can change the relative ranking within the cluster, but vertices rarely move from one cluster to another. Second, the stability is dependent on the local connections of the high ranking vertices. The network is highly stable if the high ranking vertices are connected to each other. Our findings show that the stability of a network is affected by the local properties of high centrality vertices, rather than the global properties of the entire network. Based on these local properties we can identify the stability of a network, without explicitly applying a noise model.

Original languageEnglish (US)
Title of host publicationCIKM 2016 - Proceedings of the 2016 ACM Conference on Information and Knowledge Management
PublisherAssociation for Computing Machinery
Pages2347-2352
Number of pages6
Volume24-28-October-2016
ISBN (Electronic)9781450340731
DOIs
StatePublished - Oct 24 2016
Event25th ACM International Conference on Information and Knowledge Management, CIKM 2016 - Indianapolis, United States
Duration: Oct 24 2016Oct 28 2016

Other

Other25th ACM International Conference on Information and Knowledge Management, CIKM 2016
CountryUnited States
CityIndianapolis
Period10/24/1610/28/16

Fingerprint

Centrality
Ranking
Node
Top-k
Network structure
Perturbation

Keywords

  • Betweenness
  • Closeness
  • Noise
  • Rich-club
  • Stability

ASJC Scopus subject areas

  • Business, Management and Accounting(all)
  • Decision Sciences(all)

Cite this

Ufimtsev, V., Sarkar, S., Mukherjee, A., & Bhowmick, S. (2016). Understanding stability of noisy networks through centrality measures and local connections. In CIKM 2016 - Proceedings of the 2016 ACM Conference on Information and Knowledge Management (Vol. 24-28-October-2016, pp. 2347-2352). Association for Computing Machinery. https://doi.org/10.1145/2983323.2983692

Understanding stability of noisy networks through centrality measures and local connections. / Ufimtsev, Vladimir; Sarkar, Soumya; Mukherjee, Animesh; Bhowmick, Sanjukta.

CIKM 2016 - Proceedings of the 2016 ACM Conference on Information and Knowledge Management. Vol. 24-28-October-2016 Association for Computing Machinery, 2016. p. 2347-2352.

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

Ufimtsev, V, Sarkar, S, Mukherjee, A & Bhowmick, S 2016, Understanding stability of noisy networks through centrality measures and local connections. in CIKM 2016 - Proceedings of the 2016 ACM Conference on Information and Knowledge Management. vol. 24-28-October-2016, Association for Computing Machinery, pp. 2347-2352, 25th ACM International Conference on Information and Knowledge Management, CIKM 2016, Indianapolis, United States, 10/24/16. https://doi.org/10.1145/2983323.2983692
Ufimtsev V, Sarkar S, Mukherjee A, Bhowmick S. Understanding stability of noisy networks through centrality measures and local connections. In CIKM 2016 - Proceedings of the 2016 ACM Conference on Information and Knowledge Management. Vol. 24-28-October-2016. Association for Computing Machinery. 2016. p. 2347-2352 https://doi.org/10.1145/2983323.2983692
Ufimtsev, Vladimir ; Sarkar, Soumya ; Mukherjee, Animesh ; Bhowmick, Sanjukta. / Understanding stability of noisy networks through centrality measures and local connections. CIKM 2016 - Proceedings of the 2016 ACM Conference on Information and Knowledge Management. Vol. 24-28-October-2016 Association for Computing Machinery, 2016. pp. 2347-2352
@inproceedings{690bff7886294796b374b97a58e78cb5,
title = "Understanding stability of noisy networks through centrality measures and local connections",
abstract = "Networks created from real-world data contain some inaccuracies or noise, manifested as small changes in the network structure. An important question is whether these small changes can significantly affect the analysis results. In this paper, we study the effect of noise in changing ranks of the high centrality vertices. We compare, using the Jaccard Index (JI), how many of the top-k high centrality nodes from the original network are also part of the top-k ranked nodes from the noisy network. We deem a network as stable if the JI value is high. We observe two features that affect the stability. First, the stability is dependent on the number of top-ranked vertices considered. When the vertices are ordered according to their centrality values, they group into clusters. Perturbations to the network can change the relative ranking within the cluster, but vertices rarely move from one cluster to another. Second, the stability is dependent on the local connections of the high ranking vertices. The network is highly stable if the high ranking vertices are connected to each other. Our findings show that the stability of a network is affected by the local properties of high centrality vertices, rather than the global properties of the entire network. Based on these local properties we can identify the stability of a network, without explicitly applying a noise model.",
keywords = "Betweenness, Closeness, Noise, Rich-club, Stability",
author = "Vladimir Ufimtsev and Soumya Sarkar and Animesh Mukherjee and Sanjukta Bhowmick",
year = "2016",
month = "10",
day = "24",
doi = "10.1145/2983323.2983692",
language = "English (US)",
volume = "24-28-October-2016",
pages = "2347--2352",
booktitle = "CIKM 2016 - Proceedings of the 2016 ACM Conference on Information and Knowledge Management",
publisher = "Association for Computing Machinery",

}

TY - GEN

T1 - Understanding stability of noisy networks through centrality measures and local connections

AU - Ufimtsev, Vladimir

AU - Sarkar, Soumya

AU - Mukherjee, Animesh

AU - Bhowmick, Sanjukta

PY - 2016/10/24

Y1 - 2016/10/24

N2 - Networks created from real-world data contain some inaccuracies or noise, manifested as small changes in the network structure. An important question is whether these small changes can significantly affect the analysis results. In this paper, we study the effect of noise in changing ranks of the high centrality vertices. We compare, using the Jaccard Index (JI), how many of the top-k high centrality nodes from the original network are also part of the top-k ranked nodes from the noisy network. We deem a network as stable if the JI value is high. We observe two features that affect the stability. First, the stability is dependent on the number of top-ranked vertices considered. When the vertices are ordered according to their centrality values, they group into clusters. Perturbations to the network can change the relative ranking within the cluster, but vertices rarely move from one cluster to another. Second, the stability is dependent on the local connections of the high ranking vertices. The network is highly stable if the high ranking vertices are connected to each other. Our findings show that the stability of a network is affected by the local properties of high centrality vertices, rather than the global properties of the entire network. Based on these local properties we can identify the stability of a network, without explicitly applying a noise model.

AB - Networks created from real-world data contain some inaccuracies or noise, manifested as small changes in the network structure. An important question is whether these small changes can significantly affect the analysis results. In this paper, we study the effect of noise in changing ranks of the high centrality vertices. We compare, using the Jaccard Index (JI), how many of the top-k high centrality nodes from the original network are also part of the top-k ranked nodes from the noisy network. We deem a network as stable if the JI value is high. We observe two features that affect the stability. First, the stability is dependent on the number of top-ranked vertices considered. When the vertices are ordered according to their centrality values, they group into clusters. Perturbations to the network can change the relative ranking within the cluster, but vertices rarely move from one cluster to another. Second, the stability is dependent on the local connections of the high ranking vertices. The network is highly stable if the high ranking vertices are connected to each other. Our findings show that the stability of a network is affected by the local properties of high centrality vertices, rather than the global properties of the entire network. Based on these local properties we can identify the stability of a network, without explicitly applying a noise model.

KW - Betweenness

KW - Closeness

KW - Noise

KW - Rich-club

KW - Stability

UR - http://www.scopus.com/inward/record.url?scp=84996569934&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84996569934&partnerID=8YFLogxK

U2 - 10.1145/2983323.2983692

DO - 10.1145/2983323.2983692

M3 - Conference contribution

VL - 24-28-October-2016

SP - 2347

EP - 2352

BT - CIKM 2016 - Proceedings of the 2016 ACM Conference on Information and Knowledge Management

PB - Association for Computing Machinery

ER -