Public Safety: Visualizing Crime and Emergency Response in Metro Areas

Team Members: Alex Godwin, Anand Sainath, John Stasko

    e911 Video (vimeo)
    SafePassage Video (mp4, vimeo)

    HotSketch: Drawing Police Patrol Routes among Spatiotemporal Crime Hotspots - HICSS 2017
    Exploring Spatio-Temporal Data as Personal Routes (poster) - IEEE InfoVis 2014

Public safety is an important function for any metro area. This affects policy makers, emergency personnel, and the civilian inhabitants of the city. Through ongoing research with the city of Atlanta, we are working to collect and analyze public safety data that supports emergency 911 dispatch and pedestrian safety.

Police, firefighters, and medical personnel must respond quickly to call for help from the population. Cities such as Atlanta collect data on the types of calls made to 911 centers and the details of the response. We are building on work first started in the Data Science for Social Good program, and are continuing to develop new features and displays that enable police officers and dispatch operators to understand the types of calls coming from the city.

Pedestrians do not typically have access to information on the types and frequency of crime in their neighborhoods. Atlanta releases this information publicly, but it is difficult to use the raw data to understand what crimes may affect pedestrians as they move around the city. We built a tool, Safe Passage, that helps pedestrians plan routes through the city in the context of the publicly released crime data. We provide an interface in which a person enters a route (e.g., from home to work) and reviews detailed information about that route through a combination of maps and temporal views. The person obtains details for specific points by brushing over the route, and reviews changes to the charts as the waypoints for the route are adjusted. The interface allows the user to change the filtering distance for events of interest that are close to the selected route. Temporal filtering is also provided to modify the amount of historical events included in the analysis.