Outbreaks of gastrointestinal disease occur with some frequency in North America, resulting in considerably morbidity, mortality, and cost. Outbreak detection can be improved by using simulated outbreak data to build, validate, and evaluate models that aim to improve accuracy and timeliness of outbreak detection. We have constructed a microsimulation model that depicts reasonable outbreak scenarios in space and time, and explore the use of a hidden Markov model along with supervised learning algorithms to find unique space-time outbreak signatures useful for outbreak classification.