On TV crime procedurals like Criminal Minds, analysts can scan seemingly endless databases in minutes to find very specific information about their target that reveals the perpetrator in a tidy hour-long episode. Of course, it’s much messier in real life, but replacing those TV detectives with artificial intelligence might just work.
Visual Analytics for Sense-Making in Criminal Intelligence Analysis (VALCRI) brings together siloed data, that, because it’s housed separately, can make identifying criminals difficult. VALCRI uses machine-learning technology to pore over mass amounts of data from police records and crime scene photos to suspect interviews and witness testimony. Then, like an automated Sherlock Holmes, the system can find links between disparate data points that the human brain might not so easily detect. The more information the system receives, the smarter it gets over time. “If I’m a criminal, I might usually be a house burglar, but if I see a car that’s unlocked I may burgle it, even though that’s not what I usually do,” William Wong, a computer science professor at Middlesex University, gave as an example of VALCRI’s use. “Because of this, it’s important that investigators can search across different data silos.”