PredPol’s website says this,
PredPol’s cloud-based software enables law enforcement agencies to better prevent crime in their communities by generating predictions on the places and times that future crimes are most likely to occur.
Dozens of communities across the US and overseas are experiencing dramatic reductions in crime thanks in large part to PredPol software technology.
Only three pieces of data are used to make predictions – type of crime, place of crime, and time of crime. No personal data is utilized in making these predictions.
That sounds pretty awesome but a bit too simplistic. Can’t a good cop do this with his gut? So where is this headed?
This paper discusses the advantages of folding in twitter feeds to enhance the analysis. http://ptl.sys.virginia.edu/ptl/sites/default/files/manuscript_gerber.pdf
In London, the “Datathon for Social Good” came up with a model that uses cellphone data and demographics (like ethnicity, gender, socio-economics) to get crime prediction down to a single block within a larger neighborhood. (I wouldn’t want to be sitting on a stoop on that block at the wrong time!)
The information, they say, has of course been “anonymized,” but does that give them a right to use it anymore than the TSA has a right to these pictures of us? (Yes, they still collect this raw data every time you get a body scan!)
Do they keep the data? Do they discard the identified data and keep only the anonymized? Again I ask, where is this headed?