Social Simulations: from Agent-Based Modeling to Digital Twins

arXiv cs.AI · 18 hours ago

A book chapter traces the development of social simulation from rule-based agent models to AI-enhanced systems using Large Language Models and ultimately to Social Digital Twins, which are detailed computational representations of real-world socio-technical systems. The progression moves from abstract models investigating general social mechanisms to increasingly realistic simulations of specific systems. The work examines methodological foundations, applications, advantages and limitations across each paradigm.