Berlin 2018 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

SOE: Fachverband Physik sozio-ökonomischer Systeme

SOE 22: Focus Session: Computational Social Science

SOE 22.7: Vortrag

Donnerstag, 15. März 2018, 17:45–18:00, MA 001

Towards ''Valid'' Agent-based Social Simulation — •Ingo J. Timm¹, Daniel Lebherz¹, Jan Ole Berndt¹, Simon Schmaus², Joscha Krause², and Ralf Münnich² — ¹Trier University, Trier Lab for Agent-based Simulation (TriLabS@CIRT), Trier, Germany — ²Trier University, Economics and Social Statistics,Trier, Germany

The fruitful combination of social sciences and computer science, i.e., distributed artificial intelligence, has been researched for more than two decades now: E.g., in the DFG-priority research programme on ''Socionics'' (1999-2006), social theories and mechanisms have been formalized and integrated in multiagent systems to analyse these theories and mechanisms by agent-based social simulation (ABSS) as well as to improve multiagent systems themselves. In both approaches the actors have been derived from a theoretical deliberation process.

ABSS seems to have a potential to analyse decision-making in societies in real-world situations, e.g., in health care, mobility, or environmental protection. However, this requires the reconstruction of (parts of) the population based on the socio-economic disposition of the individuals leading to the challenge: where to get sufficient data and how to validate the model? Regionalization of survey data, reconstruction of an artificial population, and simulating population dynamics in context of microsimulation is part of the expertise and research focus of the Trier University social statistics department. Consequently, we are working together on this challenge to develop an integrated approach for microsimulation and ABSS. In the talk, we will outline a first step approach for modelling and simulation of care demand.