Codifying police ‘gut’ instincts with predictive analytics
Connecting state and local government leaders
Hitachi predictive technology for law enforcement combines real-time social media and Internet data feeds with historical data.
The more data sources police and public safety officials can access, the more accurate their analyses will be – provided, of course, that they can efficiently manage the information.
To address this need, Hitachi recently announced its Visualization Predictive Crime Analytics (PCA) tool, which combines real-time social media and Internet data feeds with historical data in a visualization that can help law enforcement and first responder teams assign threat levels for every city block
PCA is part of the Hitachi Visualization Suite, a hybrid cloud-based platform that integrates real-time data and video assets from public safety systems -- such as 911 computer-aided dispatch, license plate readers, or gunshot sensors -- and presents them geospatially. It also provides geospatial visualizations for historical crime data from record management systems, social media and other sources.
This visual integration of geospatial and temporal information better illustrates the data correlations and crime patterns for law enforcement, Chris Jensen, Hitachi Data Systems Federal’s director of federal law enforcement team told GCN.
This data aggregation and analytical capability provides a valuable tool for more efficient policing and emergency response. If a shooting occurs, installed surveillance cameras can focus on that particular area as well as possible getaway routes, giving police almost immediate intelligence after an event occurs.
Law enforcement agencies need a way to filter through all the data they have to focus on what’s important, Jensen said. By looking at past crimes, call-ins for certain subjects, social media, weather and other information, PCA can help analysts surface patterns in the data and make suggestions about locations that might need more police attention.
In the past, law enforcement would devote resources to high-crime areas to quell criminal activity, but, according to Jensen, this was only a stop-gap solution, with crime returning after attention shifted elsewhere. Now law enforcement can look at patterns to uncover correlations and causations and then target resources to address the underlying problems.
The solution lets police consider factors as varied as outdoor temperature and Twitter traffic to help predict criminal activity, Jensen said. Hitachi’s interface for this disparate data aggregation is also extremely user friendly, Jensen said, which is beneficial in high stress environments. The service is icon driven, allowing one to “see as much information or as little information” as needed. It also allows for a broad overview of a particular area or dataset, but it can provide greater specificity when users click on the icons.
Hitachi Data Systems said its visualization suite 4.5 with predictive crime analytics is generally available now.
NEXT STORY: Can big data predict violent behavior?