Switched dynamic structural equation models for tracking social network topologies

Brian Baingana; Georgios B. Giannakis

doi:10.1109/GlobalSIP.2015.7418283

Switched dynamic structural equation models for tracking social network topologies

Brian Baingana, Georgios B. Giannakis

Electrical and Computer Engineering

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

4 Scopus citations

Abstract

Contagions such as the spread of popular news stories, or infectious diseases, propagate in cascades over dynamic networks with unobservable topologies. However, "social signals" such as product purchase time, or blog entry timestamps are measurable, and implicitly depend on the underlying topology, making it possible to track it over time. Interestingly, network topologies often "jump" between discrete states that may account for sudden changes in the observed signals. The present paper advocates a switched dynamic structural equation model to capture the topology-dependent cascade evolution, as well as the discrete states driving the underlying topologies. Leveraging the edge sparsity inherent to social networks, a recursive ℓ1-norm regularized least-squares estimator is put forth to jointly track the states and network topologies. A first-order proximal-gradient algorithm is developed to solve the resulting optimization problem, and numerical tests on synthetic data corroborate its efficacy.

Original language	English (US)
Title of host publication	2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	682-686
Number of pages	5
ISBN (Electronic)	9781479975914
DOIs	https://doi.org/10.1109/GlobalSIP.2015.7418283
State	Published - Feb 23 2016
Event	IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015 - Orlando, United States Duration: Dec 13 2015 → Dec 16 2015

Publication series

Name	2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015

Other

Other	IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015
Country/Territory	United States
City	Orlando
Period	12/13/15 → 12/16/15

Bibliographical note

Publisher Copyright:
© 2015 IEEE.

Keywords

Social networks
network cascade
structural equation model
switched linear systems
topology inference

Publisher link

10.1109/GlobalSIP.2015.7418283

Find It

Find It at U of M Libraries

Cite this

Baingana, B., & Giannakis, G. B. (2016). Switched dynamic structural equation models for tracking social network topologies. In 2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015 (pp. 682-686). Article 7418283 (2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/GlobalSIP.2015.7418283

Switched dynamic structural equation models for tracking social network topologies. / Baingana, Brian; Giannakis, Georgios B.
2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015. Institute of Electrical and Electronics Engineers Inc., 2016. p. 682-686 7418283 (2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015).

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

Baingana, B & Giannakis, GB 2016, Switched dynamic structural equation models for tracking social network topologies. in 2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015., 7418283, 2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015, Institute of Electrical and Electronics Engineers Inc., pp. 682-686, IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015, Orlando, United States, 12/13/15. https://doi.org/10.1109/GlobalSIP.2015.7418283

Baingana B, Giannakis GB. Switched dynamic structural equation models for tracking social network topologies. In 2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015. Institute of Electrical and Electronics Engineers Inc. 2016. p. 682-686. 7418283. (2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015). doi: 10.1109/GlobalSIP.2015.7418283

@inproceedings{e34b8b5294f149e0a3c43cc86e3d110d,

title = "Switched dynamic structural equation models for tracking social network topologies",

abstract = "Contagions such as the spread of popular news stories, or infectious diseases, propagate in cascades over dynamic networks with unobservable topologies. However, {"}social signals{"} such as product purchase time, or blog entry timestamps are measurable, and implicitly depend on the underlying topology, making it possible to track it over time. Interestingly, network topologies often {"}jump{"} between discrete states that may account for sudden changes in the observed signals. The present paper advocates a switched dynamic structural equation model to capture the topology-dependent cascade evolution, as well as the discrete states driving the underlying topologies. Leveraging the edge sparsity inherent to social networks, a recursive ℓ1-norm regularized least-squares estimator is put forth to jointly track the states and network topologies. A first-order proximal-gradient algorithm is developed to solve the resulting optimization problem, and numerical tests on synthetic data corroborate its efficacy.",

keywords = "Social networks, network cascade, structural equation model, switched linear systems, topology inference",

author = "Brian Baingana and Giannakis, {Georgios B.}",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015 ; Conference date: 13-12-2015 Through 16-12-2015",

year = "2016",

month = feb,

day = "23",

doi = "10.1109/GlobalSIP.2015.7418283",

language = "English (US)",

series = "2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015",

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

pages = "682--686",

booktitle = "2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015",

}

TY - GEN

T1 - Switched dynamic structural equation models for tracking social network topologies

AU - Baingana, Brian

AU - Giannakis, Georgios B.

PY - 2016/2/23

Y1 - 2016/2/23

N2 - Contagions such as the spread of popular news stories, or infectious diseases, propagate in cascades over dynamic networks with unobservable topologies. However, "social signals" such as product purchase time, or blog entry timestamps are measurable, and implicitly depend on the underlying topology, making it possible to track it over time. Interestingly, network topologies often "jump" between discrete states that may account for sudden changes in the observed signals. The present paper advocates a switched dynamic structural equation model to capture the topology-dependent cascade evolution, as well as the discrete states driving the underlying topologies. Leveraging the edge sparsity inherent to social networks, a recursive ℓ1-norm regularized least-squares estimator is put forth to jointly track the states and network topologies. A first-order proximal-gradient algorithm is developed to solve the resulting optimization problem, and numerical tests on synthetic data corroborate its efficacy.

AB - Contagions such as the spread of popular news stories, or infectious diseases, propagate in cascades over dynamic networks with unobservable topologies. However, "social signals" such as product purchase time, or blog entry timestamps are measurable, and implicitly depend on the underlying topology, making it possible to track it over time. Interestingly, network topologies often "jump" between discrete states that may account for sudden changes in the observed signals. The present paper advocates a switched dynamic structural equation model to capture the topology-dependent cascade evolution, as well as the discrete states driving the underlying topologies. Leveraging the edge sparsity inherent to social networks, a recursive ℓ1-norm regularized least-squares estimator is put forth to jointly track the states and network topologies. A first-order proximal-gradient algorithm is developed to solve the resulting optimization problem, and numerical tests on synthetic data corroborate its efficacy.

KW - Social networks

KW - network cascade

KW - structural equation model

KW - switched linear systems

KW - topology inference

UR - http://www.scopus.com/inward/record.url?scp=84964766281&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84964766281&partnerID=8YFLogxK

U2 - 10.1109/GlobalSIP.2015.7418283

DO - 10.1109/GlobalSIP.2015.7418283

M3 - Conference contribution

AN - SCOPUS:84964766281

T3 - 2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015

SP - 682

EP - 686

BT - 2015 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - IEEE Global Conference on Signal and Information Processing, GlobalSIP 2015

Y2 - 13 December 2015 through 16 December 2015

ER -

Switched dynamic structural equation models for tracking social network topologies

Abstract

Publication series

Other

Bibliographical note

Keywords

Publisher link

Find It

Other files and links

Fingerprint

Cite this