# American Institute of Mathematical Sciences

February  2021, 14(1): 45-76. doi: 10.3934/krm.2020048

## Opinion formation systems via deterministic particles approximation

 1 DISIM, Università degli Studi dell'Aquila, via Vetoio 1 (Coppito), 67100 L'Aquila (AQ), Italy 2 EPFL SB, Station 8, CH-1015 Lausanne, Switzerland

* Corresponding author: Simone Fagioli

Received  April 2020 Revised  August 2020 Published  February 2021 Early access  September 2020

We propose an ODE-based derivation for a generalized class of opinion formation models either for single and multiple species (followers, leaders, trolls). The approach is purely deterministic and the evolution of the single opinion is determined by the competition between two mechanisms: the opinion diffusion and the compromise process. Such deterministic approach allows to recover in the limit an aggregation/(nonlinear)diffusion system of PDEs for the macroscopic opinion densities.

Erratum: The month information has been corrected from January to February. We apologize for any inconvenience this may cause.

Citation: Simone Fagioli, Emanuela Radici. Opinion formation systems via deterministic particles approximation. Kinetic & Related Models, 2021, 14 (1) : 45-76. doi: 10.3934/krm.2020048
##### References:

show all references

##### References:
Convergence for different initial data to the stationary state (47). The left column shows the evolution in time for the reconstructed density, where the initial data are (48), (49) and (50) respectively, while the right column shows the opinions evolution in time. The (magenta) stars-line represent the mean opinion $m_1(t)$ in each case. Note that, in the second simulation, $m_1$ is not conserved in time but still converges to zero
Convergence for different initial data to the stationary state (51), where the initial data are (48), (49) and (50) respectively. Note that also in this case the mean opinion $m_1$ (star magenta line in the right column) converges to zero asymptotically
Comparison of different stationary states (46) for different values of $\alpha$. On the left the diffusion coefficient $\lambda^2 = 0.5$, on the right $\lambda^2 = 0.03$
Convergence for the initial data (48), (49) and (50) to the stationary state (53), with $\alpha = 1$. Note that the nonlinearity in the diffusion produces stationary solutions with supports that are smaller than the ones we have seen in the linear diffusion case
Stationary states in (53) for different values of $\alpha$
Opinion dynamics in presence of equally-strong leaders. Top-left initial data for followers(black), left leaders (red) and right leaders (blue). Top-right and bottom-left opinions evolutions for all the groups in short and long term-respectively. Bottom-right opinions evolutions for the followers
Opinion dynamics in presence of one strong group of leaders and one weak group of leaders. Top-left initial data for followers(black), left leaders (red) and right leaders (blue). Top-right and bottom-left opinions evolutions for all the groups in short and long term-respectively. Bottom-right opinions evolutions for the followers
Evolution of system (12). The left column concerns the case of two equally strong groups of leaders, the right column instead describes the situation where the right leader is weaker. Trolls are plotted in green and are associated to the right leaders. Top: initial data. Centre: opinions dynamic for all species. Bottom: comparison between the followers paths with or without trolls
 [1] Andrea L. Bertozzi, Dejan Slepcev. Existence and uniqueness of solutions to an aggregation equation with degenerate diffusion. Communications on Pure & Applied Analysis, 2010, 9 (6) : 1617-1637. doi: 10.3934/cpaa.2010.9.1617 [2] Laurent Desvillettes, Michèle Grillot, Philippe Grillot, Simona Mancini. Study of a degenerate reaction-diffusion system arising in particle dynamics with aggregation effects. Discrete & Continuous Dynamical Systems, 2018, 38 (9) : 4675-4692. doi: 10.3934/dcds.2018205 [3] Simone Fagioli, Yahya Jaafra. Multiple patterns formation for an aggregation/diffusion predator-prey system. Networks & Heterogeneous Media, 2021, 16 (3) : 377-411. doi: 10.3934/nhm.2021010 [4] R.A. Satnoianu, Philip K. Maini, F.S. Garduno, J.P. Armitage. Travelling waves in a nonlinear degenerate diffusion model for bacterial pattern formation. Discrete & Continuous Dynamical Systems - B, 2001, 1 (3) : 339-362. doi: 10.3934/dcdsb.2001.1.339 [5] Naoufel Ben Abdallah, Raymond El Hajj. Diffusion and guiding center approximation for particle transport in strong magnetic fields. Kinetic & Related Models, 2008, 1 (3) : 331-354. doi: 10.3934/krm.2008.1.331 [6] Sahani Pathiraja, Wilhelm Stannat. Analysis of the feedback particle filter with diffusion map based approximation of the gain. Foundations of Data Science, 2021, 3 (3) : 615-645. doi: 10.3934/fods.2021023 [7] Marco Di Francesco, Simone Fagioli, Massimiliano Daniele Rosini, Giovanni Russo. Deterministic particle approximation of the Hughes model in one space dimension. Kinetic & Related Models, 2017, 10 (1) : 215-237. doi: 10.3934/krm.2017009 [8] Antonio Suárez. A logistic equation with degenerate diffusion and Robin boundary conditions. Communications on Pure & Applied Analysis, 2008, 7 (5) : 1255-1267. doi: 10.3934/cpaa.2008.7.1255 [9] Dong Li, Xiaoyi Zhang. On a nonlocal aggregation model with nonlinear diffusion. Discrete & Continuous Dynamical Systems, 2010, 27 (1) : 301-323. doi: 10.3934/dcds.2010.27.301 [10] Yuming Paul Zhang. On a class of diffusion-aggregation equations. Discrete & Continuous Dynamical Systems, 2020, 40 (2) : 907-932. doi: 10.3934/dcds.2020066 [11] Wilhelm Stannat, Lukas Wessels. Deterministic control of stochastic reaction-diffusion equations. Evolution Equations & Control Theory, 2021, 10 (4) : 701-722. doi: 10.3934/eect.2020087 [12] Marco Caponigro, Anna Chiara Lai, Benedetto Piccoli. A nonlinear model of opinion formation on the sphere. Discrete & Continuous Dynamical Systems, 2015, 35 (9) : 4241-4268. doi: 10.3934/dcds.2015.35.4241 [13] Xiaojie Hou, Yi Li, Kenneth R. Meyer. Traveling wave solutions for a reaction diffusion equation with double degenerate nonlinearities. Discrete & Continuous Dynamical Systems, 2010, 26 (1) : 265-290. doi: 10.3934/dcds.2010.26.265 [14] Chunpeng Wang, Yanan Zhou, Runmei Du, Qiang Liu. Carleman estimate for solutions to a degenerate convection-diffusion equation. Discrete & Continuous Dynamical Systems - B, 2018, 23 (10) : 4207-4222. doi: 10.3934/dcdsb.2018133 [15] María del Mar González, Regis Monneau. Slow motion of particle systems as a limit of a reaction-diffusion equation with half-Laplacian in dimension one. Discrete & Continuous Dynamical Systems, 2012, 32 (4) : 1255-1286. doi: 10.3934/dcds.2012.32.1255 [16] Yuan Lou, Wei-Ming Ni, Shoji Yotsutani. Pattern formation in a cross-diffusion system. Discrete & Continuous Dynamical Systems, 2015, 35 (4) : 1589-1607. doi: 10.3934/dcds.2015.35.1589 [17] Maxime Breden, Christian Kuehn, Cinzia Soresina. On the influence of cross-diffusion in pattern formation. Journal of Computational Dynamics, 2021, 8 (2) : 213-240. doi: 10.3934/jcd.2021010 [18] Faustino Sánchez-Garduño, Philip K. Maini, Judith Pérez-Velázquez. A non-linear degenerate equation for direct aggregation and traveling wave dynamics. Discrete & Continuous Dynamical Systems - B, 2010, 13 (2) : 455-487. doi: 10.3934/dcdsb.2010.13.455 [19] Jakub Cupera. Diffusion approximation of neuronal models revisited. Mathematical Biosciences & Engineering, 2014, 11 (1) : 11-25. doi: 10.3934/mbe.2014.11.11 [20] Sergei Yu. Pilyugin, M. C. Campi. Opinion formation in voting processes under bounded confidence. Networks & Heterogeneous Media, 2019, 14 (3) : 617-632. doi: 10.3934/nhm.2019024

2020 Impact Factor: 1.432