Our purpose is to make realistic biosurveillance data available for researchers and to enable others to do the same. We provide tools to create realistic mimics; we also provide mimics from our own data.
We also want to improve methods for describing and simulating biosurveillance data; comments and improvements to our mimic methods are welcome and encouraged. Better mimics mean better models for biosurveillance data, which aids detection research as well.
Papers and Publications
mimics.zip: This contains mimics of 6 series (3 respiratory, 3 GI) from BioALIRT. There are 10 different mimics contained within.
mimic_0.1.tar.gz: the mimic R package, with examples, for creating mimics and inserting outbreaks. Usable on either windows or UNIX systems. The code is distributed under the GNU General Public License. Note that you can use it (or change it according to the license), but we carry no responsibility to its accuracy and use.
Links to Other Datasets
CDC EARS: 56 simulated data sets, using a negative binomial distribution with superimposed outbreaks.
- Thomas Lotze, lotze at math dot umd dot edu
- Galit Shmueli, gshmueli at rhsmith dot umd dot edu
- Inbal Yahav, iyahav at rhsmith dot umd dot edu