(Q\)-voter model with independence on signed random graphs: approximate master equations

Approximate master equations are derived for the two-state \(q\)-voter model with independence on signed random graphs, with negative and positive weights of links corresponding to antagonistic and reinforcing interactions, respectively. Depending on the mean degree of nodes, the size of the \(q\)-n...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Krawiecki, Andrzej, Gradowski, Tomasz
Format Paper
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 25.02.2024
Subjects
Approximation
Ferromagnetism
Graphs
Mathematical models
Monte Carlo simulation
Nodes
Voters
Online AccessGet full text

Cover

More Information
Summary:Approximate master equations are derived for the two-state \(q\)-voter model with independence on signed random graphs, with negative and positive weights of links corresponding to antagonistic and reinforcing interactions, respectively. Depending on the mean degree of nodes, the size of the \(q\)-neighborhood, and the fraction of the antagonistic links, with decreasing independence of agents, this model shows a first- or second-order ferromagnetic-like transition to an ordered state with one dominant opinion. Predictions of the approximate master equations concerning this transition exhibit quantitative agreement with results of Monte Carlo simulations in the whole range of parameters of the model, even if predictions of the widely used pair and mean field approximations are inaccurate. Heterogeneous pair approximation derived from the approximate master equations yields results indistinguishable from homogeneous pair approximation studied before and fails in the case of the model on networks with a small and comparable mean degree of nodes and size of the \(q\)-neighborhood.
ISSN:2331-8422