Predicting crime through pattern recognition

Predicting crime through pattern recognition

Hitachi has introduced a system that promises to predict where and when crimes are likely to occur.

According to a Fast Company article, the system, known as Hitachi’s Visualization Predictive Crime Analytics (PCA), uses a bevy of data from public safety systems, historical crime statistics, public transportation maps, weather reports and social media conversations.  It then draws on machine learning and the R statistical software to map patterns that law enforcement would otherwise miss.

The PCA is especially equipped to analyze Twitter and other forms of social media. Through a system called a latent Dirichlet allocation, the PCA can sift through every tweet tagged to a specific geography to find significant words that indicate what's happening

Hitachi’s system incorporates a visual interface complete with heat maps showing the intensity of various crime indicators as well as icons indicating guns, cell phones and surveillance cameras.  It can pinpoint an area down to 200 square meters and assign it a threat level from 0-100.

The PCA system is a hybrid-cloud platform that can run on third-party cloud platforms, including Amazon Web Services and Microsoft Azure, as well as on Hitachi's own HDS cloud system.

About half a dozen cities are already lined up to pilot the system and run double blind studies, although Hitachi declined to name the partners. Several dozen cities already use Hitachi’s video surveillance and sensor systems in their police departments.

inside gcn

  • power grid (elxeneize/Shutterstock.com)

    Electric grid protection through low-cost sensors, machine learning

Reader Comments

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above

More from 1105 Public Sector Media Group