Deploying nEmesis: Preventing Foodborne Illness by Data Mining Social Media

mined for marketing consumer goods. Starting around 2010, researchers began to realize that the same techniques could be used for influenza surveillance (Culotta 2010). Since then, social media analytics for public health has been expanded to monitor a variety of conditions, including cholera (Chunara, Andrews, and Brownstein 2012), mental health (Golder and Macy 2011), and diet (Widener and Li 2014). This body of work has shown that social media can be a useful complement to traditional methods, such as surveys of medical providers or individuals, for gathering aggregate public health statistics. Our work extends the social media analytics approach to a new domain, foodborne illness. Our most important contribution, however, is that we go beyond simply monitoring population-level prevalence. Our system, nEmesis, provides specific, actionable information, which is used to support effective public health interventions. Articles