Publication

Analysis of a nonlinear opinion dynamics model with biased assimilation

Xia, W., Ye, M., Liu, J., Cao, M. & Sun, X-M., Oct-2020, In : Automatica. 120, 8 p., 109113.

Research output: Contribution to journalArticleAcademicpeer-review

APA

Xia, W., Ye, M., Liu, J., Cao, M., & Sun, X-M. (2020). Analysis of a nonlinear opinion dynamics model with biased assimilation. Automatica, 120, [109113]. https://doi.org/10.1016/j.automatica.2020.109113

Author

Xia, Weiguo ; Ye, Mengbin ; Liu, Ji ; Cao, Ming ; Sun, Xi-Ming. / Analysis of a nonlinear opinion dynamics model with biased assimilation. In: Automatica. 2020 ; Vol. 120.

Harvard

Xia, W, Ye, M, Liu, J, Cao, M & Sun, X-M 2020, 'Analysis of a nonlinear opinion dynamics model with biased assimilation', Automatica, vol. 120, 109113. https://doi.org/10.1016/j.automatica.2020.109113

Standard

Analysis of a nonlinear opinion dynamics model with biased assimilation. / Xia, Weiguo; Ye, Mengbin; Liu, Ji; Cao, Ming; Sun, Xi-Ming.

In: Automatica, Vol. 120, 109113, 10.2020.

Research output: Contribution to journalArticleAcademicpeer-review

Vancouver

Xia W, Ye M, Liu J, Cao M, Sun X-M. Analysis of a nonlinear opinion dynamics model with biased assimilation. Automatica. 2020 Oct;120. 109113. https://doi.org/10.1016/j.automatica.2020.109113


BibTeX

@article{bfa2b7675e1c44f6bc0b790127e2e133,
title = "Analysis of a nonlinear opinion dynamics model with biased assimilation",
abstract = "This paper analyzes a nonlinear opinion dynamics model which generalizes the DeGroot model by introducing a bias parameter for each individual. The original DeGroot model is recovered when the bias parameter is equal to zero. The magnitude of this parameter reflects an individual's degree of bias when assimilating new opinions, and depending on the magnitude, an individual is said to have weak, intermediate, and strong bias. The opinions of the individuals lie between 0 and 1. It is shown that for strongly connected networks, the equilibria with all elements equal identically to the extreme value 0 or 1 is locally exponentially stable, while the equilibrium with all elements equal to the neutral consensus value of 1/2 is unstable. Regions of attraction for the extreme consensus equilibria are given. For the equilibrium consisting of both extreme values 0 and 1, which corresponds to opinion polarization according to the model, it is shown that the equilibrium is unstable for all strongly connected networks if individuals all have weak bias, becomes locally exponentially stable for complete and two-island networks if individuals all have strong bias, and its stability heavily depends on the network topology when individuals have intermediate bias. Analysis on star graphs and simulations show that additional equilibria may exist where individuals form clusters.",
keywords = "NETWORKS",
author = "Weiguo Xia and Mengbin Ye and Ji Liu and Ming Cao and Xi-Ming Sun",
year = "2020",
month = oct,
doi = "10.1016/j.automatica.2020.109113",
language = "English",
volume = "120",
journal = "Automatica",
issn = "1873-2836",
publisher = "PERGAMON-ELSEVIER SCIENCE LTD",

}

RIS

TY - JOUR

T1 - Analysis of a nonlinear opinion dynamics model with biased assimilation

AU - Xia, Weiguo

AU - Ye, Mengbin

AU - Liu, Ji

AU - Cao, Ming

AU - Sun, Xi-Ming

PY - 2020/10

Y1 - 2020/10

N2 - This paper analyzes a nonlinear opinion dynamics model which generalizes the DeGroot model by introducing a bias parameter for each individual. The original DeGroot model is recovered when the bias parameter is equal to zero. The magnitude of this parameter reflects an individual's degree of bias when assimilating new opinions, and depending on the magnitude, an individual is said to have weak, intermediate, and strong bias. The opinions of the individuals lie between 0 and 1. It is shown that for strongly connected networks, the equilibria with all elements equal identically to the extreme value 0 or 1 is locally exponentially stable, while the equilibrium with all elements equal to the neutral consensus value of 1/2 is unstable. Regions of attraction for the extreme consensus equilibria are given. For the equilibrium consisting of both extreme values 0 and 1, which corresponds to opinion polarization according to the model, it is shown that the equilibrium is unstable for all strongly connected networks if individuals all have weak bias, becomes locally exponentially stable for complete and two-island networks if individuals all have strong bias, and its stability heavily depends on the network topology when individuals have intermediate bias. Analysis on star graphs and simulations show that additional equilibria may exist where individuals form clusters.

AB - This paper analyzes a nonlinear opinion dynamics model which generalizes the DeGroot model by introducing a bias parameter for each individual. The original DeGroot model is recovered when the bias parameter is equal to zero. The magnitude of this parameter reflects an individual's degree of bias when assimilating new opinions, and depending on the magnitude, an individual is said to have weak, intermediate, and strong bias. The opinions of the individuals lie between 0 and 1. It is shown that for strongly connected networks, the equilibria with all elements equal identically to the extreme value 0 or 1 is locally exponentially stable, while the equilibrium with all elements equal to the neutral consensus value of 1/2 is unstable. Regions of attraction for the extreme consensus equilibria are given. For the equilibrium consisting of both extreme values 0 and 1, which corresponds to opinion polarization according to the model, it is shown that the equilibrium is unstable for all strongly connected networks if individuals all have weak bias, becomes locally exponentially stable for complete and two-island networks if individuals all have strong bias, and its stability heavily depends on the network topology when individuals have intermediate bias. Analysis on star graphs and simulations show that additional equilibria may exist where individuals form clusters.

KW - NETWORKS

U2 - 10.1016/j.automatica.2020.109113

DO - 10.1016/j.automatica.2020.109113

M3 - Article

VL - 120

JO - Automatica

JF - Automatica

SN - 1873-2836

M1 - 109113

ER -

ID: 133812266