Using Local Public Data Sets
After learning the secret ingredient for Flite’s Q2 hackday was big data, we spent a lot of time looking at the many public data sets available, and deciding what to incorporate in our project. One notable departure from the typical hackday format for Q2 was that we were not required to develop something specifically for the Flite platform.
This interested us, and looking at the plethora of public data sets available on the web, we kept coming back to the sets that hit closest to home: San Francisco City data.
The San Francisco Open Data Portal provides a wide range of data sets containing information related to campaign finance contribution, public transportation (MUNI), and even listings of locations for movies shot in the city.
Being somewhat typical, modern city dwellers, one thing we love about San Francisco is the many purveyors of, and availability for, tasty food and beverage. Given this preoccupation with eats, drinks, etc., one of the most interesting data sets available to us was the Health Department’s repository of Restaurant Health Inspection Scores.
We set out to build a visualization of these scores by plotting this data in an application using Google Maps.
Red sections indicate lower (high risk) health scores. Blue sections represent higher (low risk) health scores.
Our hope was to provide a heat map of the city indicating areas that have more dubious or more positive health scores. Ultimately the visualization provides more of a heat map of restaurants inspected over a period of time. Looking at the results, we came to the conclusion that health score filtering options would be necessary to accomplish what we were after.
For a fully functional visualization of health scores on a map, visit sfscores.com.
Team Asparagus (Steve Rowe, Saami Siddiqui, Omar Megdadi, Eugene Feingold, John Skinner)