Кластеризация агентов в модели ограниченного соседства
This article presents a new approach to designing agent-based bounded neighbourhood models (the Schelling’s models). An original agent-based model in the AnyLogic system has been developed, which describes the segregation processes caused by the behaviour patterns of agent-individuals. There are examined various scenarios (environment characteristics) affecting the cluster structure of the spatial distribution of agents. Using the proposed bounded neighbourhood model for two groups of agents with their individual characteristics clustering was performed. The dynamics of the segregation coefficient is obtained. Simulation results were analysed with the use of different machine learning and statistical methods.