An investigation revealed that Geolitica's predictive policing algorithm correctly predicted crime occurrences less than 1% of the time, despite police purchasing the system to prevent crime. Officials acknowledged the program failed to provide insights beyond what they already understood. While the concept of crime clustering is supported by research, the algorithm overlooks crucial factors behind crime patterns. Experts emphasize that understanding and diagnosing underlying issues is essential for effective resolution rather than mere prediction, as current methods often result in officers performing unproductive tasks instead of addressing root causes.
If you want to solve a problem, if you want to understand a problem, you need to diagnose the problem. Predictions are only proven right if you never fix the problem.
Police officials insisted that they bought the program thinking it would be helpful to prevent crime, but quickly realized it wasn't telling them anything they didn't already know.
Collection
[
|
...
]