Now artificial intelligence and machine learning mean this concept has leapt straight from the pages of science fiction into the real world. The software tries to predict where and when specific crimes will occur over the next 12 hours, and the algorithm is updated every day as new data comes in.
But can we really predict when and where a crime will take place?
Hi Ana,
My colleagues here at, BTH Sweden, are delving into this problem. Examples of their work are shown below:
Martin Boldt, Anton Borg, Martin Svensson, Jonas Hildeby, “Predicting burglars’ risk exposure and level of pre-crime preparation using crime scene data”, in Journal of Intelligent Data Analysis, volume 22, issue 1, 2018.
Anton Borg, Martin Boldt, Niklas Lavesson, Ulf Melander, Vaselka Boeva, “Detecting serial residential burglaries using clustering”, in Journal of Expert Systems with Applications, Elsevier, Vol. 44, Issue 11.
See more at:
https://www.bth.se/eng/staff/martin-boldt-mbo/
To an extend, this can be done - accuracy will depend on various factors, when you have enough data set to work with, in conjunction with ground level intelligence and applying the correct context.
If proper forensics are done , if criminals psyche is well profiled , if it is not the first time and iff (if and only if) people are dedicated in using the AI/ML to predict where it strikes again !
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Artificial Intelligence