Abstract
To track outbreaks of influenza (flu), we computed twitter rates from 31 US cities and compare rates to influenza-like-illness (ILI) surveillance rates. Over 2 flu seasons, 2012-14, significant correlations and similar graphic patterns were observed. We demonstrate an interactive dashboard \"SMART\" that allows practitioners to monitor and visualize daily changes of flu tweets and related news. Compared to regional or national approaches such a GoogleFluTrends, this system allows rapid public opinion analysis and flu outbreak detection at the local level. The SMART dashboard can provide timely and actionable information for local for agencies and practitioners.