Make this page my home page
  1. Drag the home icon in this panel and drop it onto the "house icon" in the tool bar for the browser

  2. Select "Yes" from the popup window and you're done!

Home  >  Topics  >  Crime

March 19, 2014
PrintCommentRSS

NY cops look to stop crime before it happens through data analysis

Predictive analysis — the digital dissection of police data to identify likely offenses, their locations and even perpetrators — may sound like a concept straight out of the science fiction film "Minority Report"

By Kevin Deutsch
Newsday

LONG ISLAND — Crime has plunged to record lows across much of Long Island with a new police focus on predictive crime fighting, which authorities credit with helping them stop illegal activities before they happen.

Predictive analysis  the digital dissection of police data to identify likely offenses, their locations and even perpetrators  may sound like a concept straight out of the science fiction film "Minority Report."

But it's a real system anchored in mathematical analysis, and it's transforming the way police fight crime on Long Island, New York City, and in big cities and suburbs across the country, officials and experts say.

Nassau police officials are so confident in their predictive analysis program that they replaced the department's longtime crime-mapping and statistics program, called Nass-Stat, with Strat-Com, which generates predictive information about future crimes based on intelligence and historical data.

"This is changing the way we do our jobs and the way we approach crime," said Det. Sgt. Patrick Ryder, head of the Nassau County Police department's intelligence section. "These are tools we've never had before. Now that they're here, we can actually predict what's going to happen and stop it before it does."

Since Nassau police in January began an intensified emphasis on predicting crime, police say index crime  violent offenses and property offenses  fell almost 12 percent as of March 1. The county's crime rate during those two months was the lowest on record since the department starting compiling data in 1966, police said.

In Suffolk, where predictive policing began in 2012, index crimes in 2014 fell 11 percent through Feb. 1, the latest date for which statistics were available. That represents one of the largest decreases county police have ever reported, records show.

Weather Could Be A Factor
Some of the decrease in both counties may be attributable to the biting cold and snow of January and February, police said. Studies have shown crime tends to fall when it's very cold. Still, local cops believe the recent decrease in crime is mostly a result of their tactics.

"We've really changed the way the department does business" through predictive analysis and the use of police intelligence, James Burke, chief of department for Suffolk County police, has said of the department's analytical approach.

Nassau's crime-mapping and statistics effort, Nass-Stat, which involved monthly meetings of police brass and supervisors, was modeled after the pioneering NYPD program called Comp-Stat. The NYPD initiative was credited with bringing down crime. In addition, it was notorious for top bosses aggressively questioning  and sometimes berating  commanders who failed to perform.

Police officials said Nass-Stat was more collegial but it could still be grueling for a presenter with bad crime numbers and no plan to turn them around, police said.

Strat-Com, short for strategic communication, replaced Nass-Stat, police said. Its goal is to use department resources more efficiently by sending cops to "hot spots" where a certain crime is likely to occur and pointing them toward "hot people" linked to similar crime patterns.

"We knew we needed to go in and help him [the police official in the Nass-Stat meeting], not beat him up over the problem without showing him how to fix it," said Ryder, who proposed replacing Nass-Stat. "Now, they come in and tell us what ails them, so to speak, and we write them a prescription for how we think they should attack that ailment."

One example police cite as a Strat-Com success story is the decrease in residential burglaries in neighborhoods in and near Plainview, Plainedge and Bethpage. Strat-Com identified a historical pattern of burglary spikes in the area around the same time each year and the system predicted a similar spike during that time frame this year.

Targeted Patrols
In response, more cops were dispatched to the area ahead of time to patrol, engage residents and keep an eye on recidivists with a history of committing burglaries. As a result, the annual surge never materialized, leading to a 33 percent decrease in residential burglaries in the area this year, Ryder said.

Predictive policing has also been used in New York City, where major crime is down 2 percent as of March 11, homicides fell 21 percent and shootings declined 14 percent, police said.

Despite its apparent early success, proponents of predictive policing acknowledge the technology is most effective against property crimes like break-ins and vehicle thefts, since those are the most common offenses and provide huge amounts of data for police computers to analyze. But determining when and where a murder or rape will take place is much more difficult, police said.

Predictive analysis does nothing to address or quantify the root causes of crime, meaning poverty and lack of job opportunities in a given neighborhood are not factored into Strat-Com's algorithms.

The system can identify a likely offender based on his criminal history but it cannot address his or her underlying motives, police said.

Still, police believe predictive technology will only improve, making further decreases in crime likely.

"This is the future," Ryder said.

Copyright 2014 Newsday


McClatchy-Tribune News Service

 






PoliceOne Offers