A Wavelet-based Anomaly Detector for Early Detection of Disease Outbreaks

Publication Type:

Conference Proceedings

Source:

Proceedings of the 23rd International Conference on Machine Learning (ICML), Workshop on Machine Learning Algorithms for Surveillance and Event Detection, Pittsburgh, PA (2006)

URL:

http://web.engr.oregonstate.edu/~wong/workshops/icml2006/papers/lotze.pdf

Abstract:

We describe a wavelet-based automated algorithm for
detecting disease outbreaks in temporal syndromic data.
We describe the method, which improves upon the
Goldenberg et al. (2002) algorithm and its implementation
on a diverse set of real syndromic data from multiple data
sources and multiple geographical locations. Our results
show a robust performance which is comparable to a few
recently suggested methods.