Intent-driven behavioral modeling during cross-border epidemics

Modeling real-world social situations has proven to be one of the most daunting challenges in computational social science. With the exception of simplistic, single-domain scenarios, most computational models are quickly overwhelmed with the complexity and diversity of real-world scenarios. In this paper, we apply intent-driven modeling to a complex, real-world scenario. By mapping actors'intentions to their beliefs and goals, we are able to explain their actions and propose predictions of future actions. Specifically, we look at ways to help understand and explain complex group behaviors during epidemics in relation to national borders. Using an intent-driven socio-cultural behavioral model implemented with the help of Bayesian Knowledge Bases (BKBs), we explore the actions and reactions of actors in an epidemic setting, providing insight into behaviors affecting border security. Using these tools, we are able to employ dynamic, multi-domain modeling to explain the decisions and actions taken by actors in the scenario. We validate our methodology by modeling and analyzing migration behaviors during the 2009 H1N1 pandemic in Mexico.