On the robustness of centrality measures under conditions of imperfect data

Stephen P. Borgatti; Kathleen M. Carley; David Krackhardt

doi:10.1016/j.socnet.2005.05.001

On the robustness of centrality measures under conditions of imperfect data

Stephen P. Borgatti, Kathleen M. Carley, David Krackhardt

Management

Research output: Contribution to journal › Article › peer-review

451 Citations (SciVal)

Abstract

An analysis is conducted on the robustness of measures of centrality in the face of random error in the network data. We use random networks of varying sizes and densities and subject them (separately) to four kinds of random error in varying amounts. The types of error are edge deletion, node deletion, edge addition, and node addition. The results show that the accuracy of centrality measures declines smoothly and predictably with the amount of error. This suggests that, for random networks and random error, we shall be able to construct confidence intervals around centrality scores. In addition, centrality measures were highly similar in their response to error. Dense networks were the most robust in the face of all kinds of error except edge deletion. For edge deletion, sparse networks were more accurately measured.

Original language	English
Pages (from-to)	124-136
Number of pages	13
Journal	Social Networks
Volume	28
Issue number	2
DOIs	https://doi.org/10.1016/j.socnet.2005.05.001
State	Published - May 2006

Bibliographical note

Funding Information:
This paper is part of the Dynamics Networks project in CASOS at CMU. This work was supported in part by the Department of Defense, the Office of Naval Research under grant no. 9620.1.1140071 on Dynamic Network Analysis, DARPA DAAH01-03-C-R111, and the National Science Foundation under MKIDS and the CASOS IGERT program. Additional support was provided by CASOS—the center for Computational Analysis of Social and Organizational Systems at Carnegie Mellon University ( http://www.casos.ece.cmu.edu ). The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the Department of Defense, the Office of Naval Research, Darpa, the National Science Foundation or the U.S. government.

ASJC Scopus subject areas

Anthropology
Sociology and Political Science
General Social Sciences
General Psychology

Access to Document

10.1016/j.socnet.2005.05.001

Cite this

@article{d422dca23e6f4beeb465569eea592180,

title = "On the robustness of centrality measures under conditions of imperfect data",

abstract = "An analysis is conducted on the robustness of measures of centrality in the face of random error in the network data. We use random networks of varying sizes and densities and subject them (separately) to four kinds of random error in varying amounts. The types of error are edge deletion, node deletion, edge addition, and node addition. The results show that the accuracy of centrality measures declines smoothly and predictably with the amount of error. This suggests that, for random networks and random error, we shall be able to construct confidence intervals around centrality scores. In addition, centrality measures were highly similar in their response to error. Dense networks were the most robust in the face of all kinds of error except edge deletion. For edge deletion, sparse networks were more accurately measured.",

author = "Borgatti, {Stephen P.} and Carley, {Kathleen M.} and David Krackhardt",

note = "Funding Information: This paper is part of the Dynamics Networks project in CASOS at CMU. This work was supported in part by the Department of Defense, the Office of Naval Research under grant no. 9620.1.1140071 on Dynamic Network Analysis, DARPA DAAH01-03-C-R111, and the National Science Foundation under MKIDS and the CASOS IGERT program. Additional support was provided by CASOS—the center for Computational Analysis of Social and Organizational Systems at Carnegie Mellon University ( http://www.casos.ece.cmu.edu ). The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the Department of Defense, the Office of Naval Research, Darpa, the National Science Foundation or the U.S. government.",

year = "2006",

month = may,

doi = "10.1016/j.socnet.2005.05.001",

language = "English",

volume = "28",

pages = "124--136",

number = "2",

}

TY - JOUR

T1 - On the robustness of centrality measures under conditions of imperfect data

AU - Borgatti, Stephen P.

AU - Carley, Kathleen M.

AU - Krackhardt, David

N1 - Funding Information: This paper is part of the Dynamics Networks project in CASOS at CMU. This work was supported in part by the Department of Defense, the Office of Naval Research under grant no. 9620.1.1140071 on Dynamic Network Analysis, DARPA DAAH01-03-C-R111, and the National Science Foundation under MKIDS and the CASOS IGERT program. Additional support was provided by CASOS—the center for Computational Analysis of Social and Organizational Systems at Carnegie Mellon University ( http://www.casos.ece.cmu.edu ). The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the Department of Defense, the Office of Naval Research, Darpa, the National Science Foundation or the U.S. government.

PY - 2006/5

Y1 - 2006/5

N2 - An analysis is conducted on the robustness of measures of centrality in the face of random error in the network data. We use random networks of varying sizes and densities and subject them (separately) to four kinds of random error in varying amounts. The types of error are edge deletion, node deletion, edge addition, and node addition. The results show that the accuracy of centrality measures declines smoothly and predictably with the amount of error. This suggests that, for random networks and random error, we shall be able to construct confidence intervals around centrality scores. In addition, centrality measures were highly similar in their response to error. Dense networks were the most robust in the face of all kinds of error except edge deletion. For edge deletion, sparse networks were more accurately measured.

AB - An analysis is conducted on the robustness of measures of centrality in the face of random error in the network data. We use random networks of varying sizes and densities and subject them (separately) to four kinds of random error in varying amounts. The types of error are edge deletion, node deletion, edge addition, and node addition. The results show that the accuracy of centrality measures declines smoothly and predictably with the amount of error. This suggests that, for random networks and random error, we shall be able to construct confidence intervals around centrality scores. In addition, centrality measures were highly similar in their response to error. Dense networks were the most robust in the face of all kinds of error except edge deletion. For edge deletion, sparse networks were more accurately measured.

UR - http://www.scopus.com/inward/record.url?scp=32544439820&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=32544439820&partnerID=8YFLogxK

U2 - 10.1016/j.socnet.2005.05.001

DO - 10.1016/j.socnet.2005.05.001

M3 - Article

AN - SCOPUS:32544439820

SN - 0378-8733

VL - 28

SP - 124

EP - 136

JO - Social Networks

JF - Social Networks

IS - 2

ER -

On the robustness of centrality measures under conditions of imperfect data

Abstract

Bibliographical note

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this