New Field of Predictive Policing Uses Math, Computers to Predict Crimes

Mathematicians at UCLA may be the newest crime fighters given their ground-breaking collaboration with the Los Angeles Police Department.

The researchers have come up with an algorithm that analyzes LAPD data on over 1,000 gang-related crimes and suspected gang crimes over 10 years in the Hollenbeck section of East Los Angeles, according to news reports. The neighborhood is known to be home to about 30 gangs and has seen close to 70 rivalries among gangs, news reports add.

“The goal was to bring some mathematical order to this murky, shifting landscape in Hollenbeck, where rivalries and alliances between groups are difficult to track,” according to The Los Angeles Times.

The computer model looks for “patterns and relationships between gangs,” explains news reports. Next, an attempt is made to identify which gangs were more likely to be involved with certain crimes.

In a test run, the calculations – about 80 percent of the time – identified the three gangs “most likely” to have committed a crime against a rival gang, the researchers said.

Martin Short, a co-author of the study, said the algorithm was able to ID the correct gang about half of the time. Using chance alone, led to selecting the correct gang only 17 percent of the time.

"If police believe a crime might have been committed by one of seven or eight rival gangs, our method would look at recent historical events in the area and compute probabilities as to which of these gangs are most likely to have committed crime," another author of the study, Andrea Bertozzi, who is the director of applied mathematics at UCLA, said in a statement carried by TMCnet.

The model can be used to investigate other types of crime, such as computer hackers or finding members in a terrorist cell, The Times said. The new field is known as “predictive policing,” The Times said.

In addition, the model can be used in marketing efforts. PhysOrg.com noted that businesses could direct their ads to “consumers who would be most interested in their products” by simply knowing their “shopping behavior.”

In its analysis, PhysOrg.com says the UCLA research “represents the first scholarly study of gang violence of its kind.”

The lead author of the study was Alexey Stomakhin, a UCLA doctoral student in applied mathematics, PhysOrg.com said.