Weaving multi-agent modeling and big data for stochastic process inference

In this paper, we develop a stochastic process tool to tell the stories behind big data with agent-based models. Specifically, we identify an agent-based model as a stochastic process that generates the big data, and make inferences by solving the agent-based model under the constraint of the data. We hope to use this tool to create a bridge between those who have access to big data and those who use agent-based simulators to convey their insight about these data.