Model of public opinion dynamics in modern society

Abstract

Nowadays more and more scientific interest is paid to social processes. It is relatively new, but very important direction because knowledge in this area might have significant impact in many fields of modern life. The purpose of this article is to create a model of public opinion dynamics in modern society. The model is iterative and considers a number of features which are typical for modern society. One of the key components of the model is political parties. They have significant influence on the public opinion. The model takes into account the impact of mass media. It considers both independent and controlled by a political party mass media. Also the model considers influence of social networks, messengers and telephony. In addition, it takes into account geographical position of each state, length of common border between states or social groups and possibly different density of population in each of the states.

The model is developed with computer implementation in mind, which significantly simplifies further steps. The author of the article implemented the model in Python programming language with help of tkinter library for user interface and matplotlib for results displaying. Also he gives some implementation advice and performance optimizations. The implemented model has flexible and easy to understand and edit configuration, which makes using of it very convenient.

In the last part of the article the built model was tested against several cases which results could be predicted. The first case shows debate of two political parties where first one has notable advantage. The second case shows polarization of the simulated society, where each ideology dominates in a separate region. The last case shows winning of a political party which has significant control over almost all mass media in the simulated society. After simulations of the cases, the model showed results similar to what was expected which indicates that the created in this article model is correct.