Event

Location: Room 2018, 1880 Pratt Drive, Corporate Research Center
Time: 2:30 PM

Speaker 1: Nidhi Parikh

Title: Modeling Human Behavior in the Aftermath of a Hypothetical Improvised Nuclear Detonation

Abstract: We focus on modeling human behavior in the aftermath of an improvised nuclear detonation in Washington DC. Earlier studies of this scenario have focused on modeling physical impact or evaluating sheltering vs. evacuation strategies, ignoring human behavioral response to the event like family members looking for each other, survivors helping others, and so on. We model human behavior in the aftermath of the event and its dependency with infrastructural systems using framework of options. Six different behavioral options are modeled: household reconstitution, evacuation, healthcare-seeking, worry, shelter-seeking, and aiding & assisting others. Agent decision-making takes into account their health status, information about family members, information about the event, and their local environment. We combine these behavioral options into five different behavior models of increasing complexity and show that as we add more behaviors outcomes change non-monotonically which means one should be careful about including all relevant behaviors. Modeling human behavior could also help align response policy with survivors' behavior, for example, we show that restoring communication can have a positive impact on health.

Speaker 2: S M Shamimul Hasan

Title: Linked Data Access Framework for Computational Networked Epidemiology Datasets

Abstract: Computational epidemiology is a data-driven life science discipline that uses modeling and simulation to develop effective strategies for containment of infectious diseases. Being a data-driven science, it produces and consumes massively large datasets. But akin to life sciences fields, the nature and types of data involved are in constant flux. These characteristics together present a unique challenge in terms of size, dynamic nature, and volume of data. We develop a federation of network based simulation and analysis for computational epidemiology datasets and provide a simplified (homogeneous) querying framework. We show that the proposed data access framework can hide schema and location heterogeneity while efficiently addressing queries that can span the entire computational epidemiology pipeline: from model construction to simulation output.