|
|
|
|
|
|
by goostavos
4793 days ago
|
|
|
Seems like a system that's destined to be rife with false positives. I can only see this data being misused by lawyers and law enforcement to support a bad assumption. Anyone with a basic familiarity with statistics knows misinterpreting is remarkably easy to do; misrepresenting it is even easier. I'd hate to have a coincidental pattern become "proof" in some kind of court case. |
|
|
Especially when the pattern is what the cops use to identify the suspect in the first place.
There is a real problem with big data in the courtroom. Take the intersection of a victim, a crime and a suspect and look at a million data points and you'll find several thousand that match. Put only those in front of a jury and you have something that looks like a mountain of evidence but is really just systematically applying confirmation bias to random noise.