POLICE Magazine

OCT 2018

Magazine for police and law enforcement

Issue link: https://policemag.epubxp.com/i/1037204

Contents of this Issue


Page 68 of 80

POLICE AGENCIES ACROSS AMERICA are being asked to do an increasing number of things at an increased frequency despite budgets that have either atlined or been cut. Agencies are seeing increases in the numbers of calls for service while at the same time dealing with a shrinking number of ocers on the streets. Consequently, the watch word at departments of every size is "eciency." One of the ways to become more ecient is to deploy ocers to the places most likely to have incidences of crime occur. •is is not exactly a new concept. Agencies have for many years had maps of their jurisdictions posted to cork-boards in which pins of varying colors—signifying di€erent types of crimes—were placed so that ocers could get a quick visualization of where they should focus their e€orts. But with the advent of artiƒcial intelligence and machine learning, the concept of crime mapping has evolved consid- erably, leading to the rapid emergence of highly sophisticated predictive policing so„ware that not only looks at past crimes, but at massive volumes of other types of data that can indicate the possibility of future criminal activity. PREDICTING CRIMINAL ACTIVITY £ Using sophisticated so„ware, law enforcement agencies can now leverage the power of "big data" to glean a much deeper un- derstanding of the predictors of criminal activity. Motorola Solutions' CommandCentral analytics so„ware includes predictive analysis technology. •is technology pro- vides police departments with a visual map of predictive crime areas that are a thousand by a thousand square feet. •ese boxes include tactical area information for ocers on patrol. Crime area boxes are ranked by their likelihood of a crime oc- curring during a given shi„. Police departments can view a list of predictive areas sorted by priority. •ey can drill down to view police report details of the most recent crimes that hap- pened in these areas. DJ Seals, a former police detective and an expert on Com- mandCentral predictive technology, says predictive technolo- gies need to go beyond hot spotting maps that are focused on a few weeks of crime data. "•ey need to have forward-looking or machine-learn- ing mathematics that take into account historical patterns of crimes," Seals explains. "A historical view will provide insights into crime patterns throughout the year, including seasonal crimes during holidays." Seals says that by bringing historical and recent crime pat- terns together, predictive technologies provide a more accurate look at what may occur in the coming days, or even the coming patrol shi„s. "One of the most important pieces of information that true predictive technologies can o€er patrol ocers is a breakdown of predictions by patrol shi„s," Seals says. "Di€erent shi„s throughout the day typically see di€erent types of crimes. At- tempting to predict a day in its entirety vs. focusing on shi„s can lead to inaccurate information." Seals uses the example of a police agency experiencing regular robberies at a speciƒc store where predictive analyt- ics showed a high recurrence of the„ at that location. "•e agency began looking into the details of the crimes," Seals says. "A common theme was that the suspect le„ the store, ran to the north, and cut through a strip mall. Behind that strip mall was a path through the woods that led to residences." When the next call came in, ocers not only went to the store, they also set up ocers by the path behind the strip mall. "•ey apprehended the suspect who had an armful of stolen goods. Using predictive technology intelligence, they were able to identify a pattern that led them to capture an elusive crimi- nal and stop the recurring the„s at this store," Seals says. MORE THAN NUMBERS £ Seals points out that predictive technologies must also go be- yond dots and boxes on a map. Of course, predictive policing so„ware looks at historical crime data—that goes without saying—but today's so„ware 8 | SP E C I A L R E P O RT | I N V E S T IG AT I V E T E C H NO L O G I E S PHOTO: MOTOROLA DOUG WYLLIE FIGHTING CRIME WITH PREDICTIVE POLICING USING SOPHISTICATED SOFTWARE, LAW ENFORCEMENT AGENCIES CAN NOW LEVERAGE THE POWER OF "BIG DATA" TO GLEAN A MUCH DEEPER UNDERSTANDING OF THE PREDICTORS OF CRIMINAL ACTIVITY. £

Articles in this issue

Archives of this issue

view archives of POLICE Magazine - OCT 2018