How Safe is Greensboro?

After moving to Greensboro, I got chatting with a friend about crime in the city. Both of us knew that crime is an issue for Greensboro -- it has a crime rate on par with the national average -- but weren't sure how that played out in the different neighborhoods.

It turns out that you can query the Greensboro Police Department's database for incidents that occurred within a certain distance of a given address in a given date range. Likewise, there's Spot Crime, which will give you a map of incidents over a date range.

These are useful, but don't allow for actual statistical analysis. If you're interested in analyzing crime patterns, you need more than anecdotal data: the fact that someone got mugged in your neighborhood doesn't mean the neighborhood is unsafe; the fact that someone gets mugged twice a week all year long, does. It occurred to me that it should be possible to query the GPD database automatically and assemble the data over time.

I wrote a script in Python to query the system for a series of addresses from 2006 to the present and store the data in MongoDB to make it available for analysis. Over the next month, I plan to post the source code and the data I've collected on github as well as more explanation about how the querying system works.

In the meantime, here are simple examples of things that one can do with this data. Using the Gheat, it's possible to create heat-maps of burglaries that occurred in the city in 2008, 2009, and 2010. With Google Fusion Tables, it's also possible to generate maps capturing the details of all burglaries. I've included a set of paired maps below: the first being a heat map and the second, a fusion tables map. If you click on the dots on the fusion table map, you'll see details for the incident.

2008 - Burglaries in Central Greensboro

2009 - Burglaries in Central Greensboro

2010 - Burglaries in Central Greensboro