Fighting Violent Gang Crime With Math

A formula can narrow down suspects in a gang shooting without knowing their identifies, the victim, or even the location. The key is the mathematical patterns behind human behavior.

UCLA mathematicians analyzed more than 1,000 gang crimes and suspected gang crimes--about half of them unsolved--over a 10-year period in an East Los Angeles police district known as Hollenbeck, home to some 30 gangs, and almost twice as many gang rivalries. The researchers assessed the network of gangs, and their interactions, to deduce the most likely players in any given event based on past behavior and relationships. As it turns out, it works.

"Our algorithm placed the correct gang rivalry within the top three most likely rivalries 80 percent of the time, which is significantly better than chance," said Martin Short, a UCLA adjunct assistant professor of mathematics and co-author of the study published in the journal Inverse Problems. "That narrows it down quite a bit, and that is when we don’t know anything about the crime victim or perpetrator."

Investigators will still need to hit the streets. They need evidence, not an algorithm, for court. But after an unsolved shooting, it might be easier to start that investigation with the gangs the algorithim says are mostly likely to be involved. The approach could save considerable time and resources. The correct gang was ranked first--rather than just among the top three--50 percent of the time, further narrowing down suspects when identities and other clues are not known.

Human behavior, only observable in databases, may be the next smoking gun.