%0 Journal Article
%T Extracting Data from Disparate Sources for Agent-Based Disease Spread Models
%A M. Laskowski
%A B. C. P. Demianyk
%A J. Benavides
%A M. R. Friesen
%A R. D. McLeod
%A S. N. Mukhi
%A M. Crowley
%J Epidemiology Research International
%D 2012
%I Hindawi Publishing Corporation
%R 10.1155/2012/716072
%X This paper presents a review and evaluation of real data sources relative to their role and applicability in an agent-based model (ABM) simulating respiratory infection spread a large geographic area. The ABM is a spatial-temporal model inclusive of behavior and interaction patterns between individual agents. The agent behaviours in the model (movements and interactions) are fed by census/demographic data, integrated with real data from a telecommunication service provider (cellular records), traffic survey data, as well as person-person contact data obtained via a custom 3G smartphone application that logs Bluetooth connectivity between devices. Each source provides data of varying type and granularity, thereby enhancing the robustness of the model. The work demonstrates opportunities in data mining and fusion and the role of data in calibrating and validating ABMs. The data become real-world inputs into susceptible-exposed-infected-recovered (SEIR) disease spread models and their variants, thereby building credible and nonintrusive models to qualitatively model public health interventions at the population level. 1. Introduction Complex networks underlie the transmission dynamics of many epidemiological models of disease spread, in particular agent-based models (ABMs). Network-based epidemiological models use a percolation-like principle to simulate disease spread through the population [1], and there are a large number of studies on ABMs and network-based epidemiological models. Agent-based models are of increasing interest due to their potential to capture complex emergent behaviours during the course of a simulated epidemic, where these behaviours arise from the nonlinearities of human-human contacts [2]. ABMs may employ an explicit or implicit social contact network defined by structured agent interactions. In the explicit case, a disease model (e.g., susceptible-exposed-infected-recovered or SEIR type) can be implemented directly on the network. In the case of ABM, these resemble simulation models rather than the steady-state analysis of network-based models mentioned in [1]. In all cases, though, the fidelity of the agent-based framework (model) relies in part on the credibility of the social contact network data that feeds it, defining agentsĄŻ characteristics, behaviours, and interactions within the model. Potential data sources to define agents include census and demographic data (coarse) and finer-grained data made available by various means of polling personal electronics such as cell phones. In related work it was demonstrated that data to
%U http://www.hindawi.com/journals/eri/2012/716072/