Differentially Private Generation of Social Networks via Exponential Random Graph Models

Fang Liu; Evercita Eugenio; Ick Hoon Jin; Claire Bowen

doi:10.1109/COMPSAC48688.2020.00-11

Differentially Private Generation of Social Networks via Exponential Random Graph Models

Fang Liu, Evercita Eugenio, Ick Hoon Jin, Claire Bowen

Department of Applied Statistics

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

6 Citations (Scopus)

Abstract

Many social networks contain sensitive relational information. One approach to protect the sensitive relational information while offering flexibility for social network research and analysis is to release synthetic social networks at a pre-specified privacy risk level, given the original observed network. We propose the DP-ERGM procedure that synthesizes networks that satisfy the differential privacy (DP) via the exponential random graph model (EGRM). We apply DP-ERGM to a college student friendship network and compare its original network information preservation in the generated private networks with two other approaches: differentially private DyadWise Randomized Response (DWRR) and Sanitization of the Conditional probability of Edge given Attribute classes (SCEA). The results suggest that DP-EGRM preserves the original information significantly better than DWRR and SCEA in both network statistics and inferences from ERGMs and latent space models. In addition, DP-ERGM satisfies the node DP, a stronger notion of privacy than the edge DP that DWRR and SCEA satisfy.

Original language	English
Title of host publication	Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020
Editors	W. K. Chan, Bill Claycomb, Hiroki Takakura, Ji-Jiang Yang, Yuuichi Teranishi, Dave Towey, Sergio Segura, Hossain Shahriar, Sorel Reisman, Sheikh Iqbal Ahamed
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1695-1700
Number of pages	6
ISBN (Electronic)	9781728173030
DOIs	https://doi.org/10.1109/COMPSAC48688.2020.00-11
Publication status	Published - 2020 Jul
Event	44th IEEE Annual Computers, Software, and Applications Conference, COMPSAC 2020 - Virtual, Madrid, Spain Duration: 2020 Jul 13 → 2020 Jul 17

Publication series

Name	Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020

Conference

Conference	44th IEEE Annual Computers, Software, and Applications Conference, COMPSAC 2020
Country/Territory	Spain
City	Virtual, Madrid
Period	20/7/13 → 20/7/17

Bibliographical note

Funding Information:
Fang Liu and Evercita C. Eugenio are/were supported by the National Science Foundation (NSF) Grants #1546373 and #1717417. Ick Hoon Jin was partially supported by the Yonsei University Research Fund of 2019-22-0210 and National Research Foundation of Korea (NRF 2020R1A2C1A01009881), and Claire Bowen was supported by the NSF Graduate Research Fellowship # DGE-1313583 during part of the paper’s development. The publication has been assigned the Sandia National Laboratories identifier SAND2020-5256 C. We thank Dr. Johnny Zhiyong Zhang from the Lab for Big Data Methodology in the Department of Psychology at the University of Notre Dame for providing the Chinese college friendship data.

Publisher Copyright:
© 2020 IEEE.

All Science Journal Classification (ASJC) codes

Artificial Intelligence
Computer Networks and Communications
Computer Science Applications
Hardware and Architecture
Software
Education

Access to Document

10.1109/COMPSAC48688.2020.00-11

Cite this

Liu, F., Eugenio, E., Jin, I. H., & Bowen, C. (2020). Differentially Private Generation of Social Networks via Exponential Random Graph Models. In W. K. Chan, B. Claycomb, H. Takakura, J-J. Yang, Y. Teranishi, D. Towey, S. Segura, H. Shahriar, S. Reisman, & S. I. Ahamed (Eds.), Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020 (pp. 1695-1700). Article 9202793 (Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/COMPSAC48688.2020.00-11

Liu, Fang ; Eugenio, Evercita ; Jin, Ick Hoon et al. / Differentially Private Generation of Social Networks via Exponential Random Graph Models. Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020. editor / W. K. Chan ; Bill Claycomb ; Hiroki Takakura ; Ji-Jiang Yang ; Yuuichi Teranishi ; Dave Towey ; Sergio Segura ; Hossain Shahriar ; Sorel Reisman ; Sheikh Iqbal Ahamed. Institute of Electrical and Electronics Engineers Inc., 2020. pp. 1695-1700 (Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020).

@inproceedings{9034b2924b924f52bfc364c9e8d11ca9,

title = "Differentially Private Generation of Social Networks via Exponential Random Graph Models",

abstract = "Many social networks contain sensitive relational information. One approach to protect the sensitive relational information while offering flexibility for social network research and analysis is to release synthetic social networks at a pre-specified privacy risk level, given the original observed network. We propose the DP-ERGM procedure that synthesizes networks that satisfy the differential privacy (DP) via the exponential random graph model (EGRM). We apply DP-ERGM to a college student friendship network and compare its original network information preservation in the generated private networks with two other approaches: differentially private DyadWise Randomized Response (DWRR) and Sanitization of the Conditional probability of Edge given Attribute classes (SCEA). The results suggest that DP-EGRM preserves the original information significantly better than DWRR and SCEA in both network statistics and inferences from ERGMs and latent space models. In addition, DP-ERGM satisfies the node DP, a stronger notion of privacy than the edge DP that DWRR and SCEA satisfy.",

author = "Fang Liu and Evercita Eugenio and Jin, {Ick Hoon} and Claire Bowen",

note = "Funding Information: Fang Liu and Evercita C. Eugenio are/were supported by the National Science Foundation (NSF) Grants #1546373 and #1717417. Ick Hoon Jin was partially supported by the Yonsei University Research Fund of 2019-22-0210 and National Research Foundation of Korea (NRF 2020R1A2C1A01009881), and Claire Bowen was supported by the NSF Graduate Research Fellowship # DGE-1313583 during part of the paper{\textquoteright}s development. The publication has been assigned the Sandia National Laboratories identifier SAND2020-5256 C. We thank Dr. Johnny Zhiyong Zhang from the Lab for Big Data Methodology in the Department of Psychology at the University of Notre Dame for providing the Chinese college friendship data. Publisher Copyright: {\textcopyright} 2020 IEEE.; 44th IEEE Annual Computers, Software, and Applications Conference, COMPSAC 2020 ; Conference date: 13-07-2020 Through 17-07-2020",

year = "2020",

month = jul,

doi = "10.1109/COMPSAC48688.2020.00-11",

language = "English",

series = "Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1695--1700",

editor = "Chan, {W. K.} and Bill Claycomb and Hiroki Takakura and Ji-Jiang Yang and Yuuichi Teranishi and Dave Towey and Sergio Segura and Hossain Shahriar and Sorel Reisman and Ahamed, {Sheikh Iqbal}",

booktitle = "Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020",

address = "United States",

}

Liu, F, Eugenio, E, Jin, IH & Bowen, C 2020, Differentially Private Generation of Social Networks via Exponential Random Graph Models. in WK Chan, B Claycomb, H Takakura, J-J Yang, Y Teranishi, D Towey, S Segura, H Shahriar, S Reisman & SI Ahamed (eds), Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020., 9202793, Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020, Institute of Electrical and Electronics Engineers Inc., pp. 1695-1700, 44th IEEE Annual Computers, Software, and Applications Conference, COMPSAC 2020, Virtual, Madrid, Spain, 20/7/13. https://doi.org/10.1109/COMPSAC48688.2020.00-11

Differentially Private Generation of Social Networks via Exponential Random Graph Models. / Liu, Fang; Eugenio, Evercita; Jin, Ick Hoon et al.
Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020. ed. / W. K. Chan; Bill Claycomb; Hiroki Takakura; Ji-Jiang Yang; Yuuichi Teranishi; Dave Towey; Sergio Segura; Hossain Shahriar; Sorel Reisman; Sheikh Iqbal Ahamed. Institute of Electrical and Electronics Engineers Inc., 2020. p. 1695-1700 9202793 (Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020).

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

TY - GEN

T1 - Differentially Private Generation of Social Networks via Exponential Random Graph Models

AU - Liu, Fang

AU - Eugenio, Evercita

AU - Jin, Ick Hoon

AU - Bowen, Claire

N1 - Funding Information: Fang Liu and Evercita C. Eugenio are/were supported by the National Science Foundation (NSF) Grants #1546373 and #1717417. Ick Hoon Jin was partially supported by the Yonsei University Research Fund of 2019-22-0210 and National Research Foundation of Korea (NRF 2020R1A2C1A01009881), and Claire Bowen was supported by the NSF Graduate Research Fellowship # DGE-1313583 during part of the paper’s development. The publication has been assigned the Sandia National Laboratories identifier SAND2020-5256 C. We thank Dr. Johnny Zhiyong Zhang from the Lab for Big Data Methodology in the Department of Psychology at the University of Notre Dame for providing the Chinese college friendship data. Publisher Copyright: © 2020 IEEE.

PY - 2020/7

Y1 - 2020/7

N2 - Many social networks contain sensitive relational information. One approach to protect the sensitive relational information while offering flexibility for social network research and analysis is to release synthetic social networks at a pre-specified privacy risk level, given the original observed network. We propose the DP-ERGM procedure that synthesizes networks that satisfy the differential privacy (DP) via the exponential random graph model (EGRM). We apply DP-ERGM to a college student friendship network and compare its original network information preservation in the generated private networks with two other approaches: differentially private DyadWise Randomized Response (DWRR) and Sanitization of the Conditional probability of Edge given Attribute classes (SCEA). The results suggest that DP-EGRM preserves the original information significantly better than DWRR and SCEA in both network statistics and inferences from ERGMs and latent space models. In addition, DP-ERGM satisfies the node DP, a stronger notion of privacy than the edge DP that DWRR and SCEA satisfy.

AB - Many social networks contain sensitive relational information. One approach to protect the sensitive relational information while offering flexibility for social network research and analysis is to release synthetic social networks at a pre-specified privacy risk level, given the original observed network. We propose the DP-ERGM procedure that synthesizes networks that satisfy the differential privacy (DP) via the exponential random graph model (EGRM). We apply DP-ERGM to a college student friendship network and compare its original network information preservation in the generated private networks with two other approaches: differentially private DyadWise Randomized Response (DWRR) and Sanitization of the Conditional probability of Edge given Attribute classes (SCEA). The results suggest that DP-EGRM preserves the original information significantly better than DWRR and SCEA in both network statistics and inferences from ERGMs and latent space models. In addition, DP-ERGM satisfies the node DP, a stronger notion of privacy than the edge DP that DWRR and SCEA satisfy.

UR - http://www.scopus.com/inward/record.url?scp=85094131386&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85094131386&partnerID=8YFLogxK

U2 - 10.1109/COMPSAC48688.2020.00-11

DO - 10.1109/COMPSAC48688.2020.00-11

M3 - Conference contribution

AN - SCOPUS:85094131386

T3 - Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020

SP - 1695

EP - 1700

BT - Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020

A2 - Chan, W. K.

A2 - Claycomb, Bill

A2 - Takakura, Hiroki

A2 - Yang, Ji-Jiang

A2 - Teranishi, Yuuichi

A2 - Towey, Dave

A2 - Segura, Sergio

A2 - Shahriar, Hossain

A2 - Reisman, Sorel

A2 - Ahamed, Sheikh Iqbal

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 44th IEEE Annual Computers, Software, and Applications Conference, COMPSAC 2020

Y2 - 13 July 2020 through 17 July 2020

ER -

Liu F, Eugenio E, Jin IH, Bowen C. Differentially Private Generation of Social Networks via Exponential Random Graph Models. In Chan WK, Claycomb B, Takakura H, Yang J-J, Teranishi Y, Towey D, Segura S, Shahriar H, Reisman S, Ahamed SI, editors, Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020. Institute of Electrical and Electronics Engineers Inc. 2020. p. 1695-1700. 9202793. (Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020). doi: 10.1109/COMPSAC48688.2020.00-11

Differentially Private Generation of Social Networks via Exponential Random Graph Models

Abstract

Publication series

Conference

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this