Modeling Social Media Effect on Civil Revolutions
Introduction
Agent-based modeling ( ABM ) helps us build an artificial environment for modeling problems with complicated behaviors . This simulated environment can refer to the society , nature , and any type of relationships among their agents , which have real meaning . For example , we can define people , viruses , cars , and roads each as an agent in the ABM . Users decide on agent ’ s behaviors , agent ’ s degree of rationalities , and rules of agent ’ s interactions with other agents , who have memory , can learn , and can evolve .
In ABM , there are micro- and macro-level dynamics . Each agent has its own attributes to decide on its actions , which create interactions in both space and time ( Miller & Page , 2004 ). The agent ’ s interactions have a one to one relationship in the model and create micro-level dynamics . Agent-based interactions might lead to a global emergent behavior that can affect the macro level dynamics , such as a social context with socio-economic conditions , power dynamics in the country , and political situations . These macro-level dynamics also have an effect on micro-level dynamics . It can be fair to call this a “ bottom-up approach ” meaning interactions happening
at the bottom level influence the interactions at the top .
We apply the system dynamics approach to understand the big picture of the problem ( SD ). This approach helps us to link subparts of the problem and their relationships to see the whole dynamics . Each arrow represents causal relationships that mean a variable in the beginning of the arrow has an influence on other variable in the end of the arrow ( Sterman , 2000 ). This influence can be positive (+) or negative ( − ). Positive sign means that an increase in the variable at the beginning of the arrow can lead to an increase in the variable at the end of the arrow . On the other hand , negative sign indicates that an increase in the variable at the beginning of the arrow can lead to a decrease in the variable at the end of the arrow . If there is an ambiguity in relationships , we leave the arrow ’ s sign blank .
There is a feedback loop dynamics between micro- and macro-level relationships . Figure 1 shows the feedback dynamics to explain why we apply ABM and SD approaches together to work on our model . Micro-level dynamics can create emergent behaviors , which develop macro-level dynamics to influence agent ’ s interactions .
Figure 1 . Feedback loop between micro- and macro-level dynamics of the ABM .
16