Simulation of Multivariate Spatial-Temporal Outbreak Data for Detection Algorithm Evaluation

Authors:
Min Zhang;Xiaohui Kong;Garrick L. Wallstrom
Affiliations:
Department of Biomedical Informatics, University of Pittsburgh, Pittsburgh, 15260;Department of Biomedical Informatics, University of Pittsburgh, Pittsburgh, 15260;Department of Biomedical Informatics, University of Pittsburgh, Pittsburgh, 15260
Venue:
BioSecure '08 Proceedings of the 2008 International Workshop on Biosurveillance and Biosecurity
Year:
2008

Citing 1
Cited 0

Detection of emerging space-time clusters

Proceedings of the eleventh ACM SIGKDD international conference on Knowledge discovery in data mining

Quantified Score

Hi-index	0.03

Visualization

Abstract

We developed a template-driven spatial-temporal multivariate outbreak simulator that can generate multiple data streams of outbreak data for evaluating detection algorithms used in disease surveillance systems. The simulator is controlled via intuitive parameters that describe features of the outbreak and surveillance system such as the elevated risk of disease, surveillance data coverage, case behavior probabilities, and the distribution of behavior times. We provide examples of temporal and spatial-temporal outbreak simulations. Our simulator is a useful tool for evaluating of outbreak detection algorithms.