Estimating Individual Behaviour from Massive Social Data for an Urban Agent-Based Model

This paper presents the most recent developments in an on-going programme of work towards a
realistic agent-based model of urban dynamics. The focus of the paper is on the development of
a framework for calibrating an agent-based model of urban dynamics using novel data from the
Twitter social media service. In particular, we discuss initial attempts to elucidate information
about peoples' daily spatio-temporal behaviour and how such insight can be used for the benefit of
agent-based models. The ultimate aim of the modelling work is to better understand the spatio-
temporal movement patterns within the city.