The Study Of Crowd Behavior Evolution With Emotional Contagion In Political Rallies

In recent years,typical political rallies always attract numerous people to gather together,such as the U.S president election and South Korea's parade to protest President Park Geun-hye.Consequently,large-scale group conflicts are becoming severe potential threats to public security,especially for people with different political beliefs.Through modeling and simulating the evolution of crowd behavior in political rallies,we can accurately capture the changing trend of crowd movement in advance,and thus implement effective monitoring and intervention in time.Thereby,it can drastically reduce the potential risk of conflicts to help maintain social stability.However,most of the existing work focus on scenarios such as mob riots or emergent accidents,and the behaviors of a crowd with one certain emotion.As a result,these methods cannot capture all the characteristics that a political rally concerns.In order to solve above problems,this paper comprehensively considers the psychological and movement state of individuals during the crowd movement process.To be more specific,we reflect the psychological state of the crowd in the individual's movement behavior in real time through the mechanism of emotional infection,and propose two group behavior simulation methods for the political rally scene.The first one is a crowd simulation method based on SIR model.Specifically,we first analyze the typical political rally scenes and model the individuals in the scene,including the individual emotions,political opinions,movement behaviors,and personalities.The "extroversion" and “empathy” factors are selected in the OCEAN personality model as the main personality factors affecting individuals in political rallies.Based on this,we combine the SIR model and individual personality factors and propose an emotional infection model which is used to describe the process of emotional spread among people with different political perspectives.Trough combining the individual's direction of emotional infection and their original movement direction,we use the social force model to update the individual's final movement direction and achieve the simulation of crowd behaviors under political rally scenarios.Aiming at the problem that the individual personality in the first method is insufficient to fully describe the individuals' emotional states,we further propose a more comprehensive crowd simulation method based on the PAD model.Specifically,we first use the OCEAN model and the OCC model to analyze individuals' personalities and emotions in a political rally.Then we use the PAD model to integrate individual personality and emotions to model the individual emotional states.Then,we judge whether the individual is affected by emotional infections according to his real-time emotional states and emotion impact of his surrounding peers.Finally,the cellular automata model is used to model the individual behaviors in microscopic,with the help of the fluid model,we can constrain individual behaviors by categorize individuals with similar tendency of movement into the same class.Based on the methods described above,the individual's movement status in the group is obtained.The experimental results show that the two crowd simulation methods combined with emotional infection proposed in this paper can intuitively and accurately represent the emotional changes of individuals in political rally scenes,and reflect the dynamic evolution process of individual behaviors and crowd behaviors in political rally scenarios effectively.