New Algorithm Could Catch Serial Killers

A former investigative reporter is trying to change the way law enforcement tracks serial killers. Thomas Hargrove has created The Murder Accountability Project – “an open-source website that uses vast banks of data from state, local and FBI sources to provide insight into murders, both unsolved and solved,” according to IFLScience.com. According to their website, this site contains “the most complete accounting of homicides available anywhere.”

After examining in depth statistics surrounding prostitution “I asked myself, ‘Do you suppose it’s possible to teach a computer how to spot serial killers?'” Hargrove told Bloomberg.  He then proceeded to use this raw data to connect 14 unsolved murders in Gary, Indiana. 

Hargrove and his team “attempted to develop an algorithm that could look for common features and links between previously unassociated murders,” IFLScience reported. He tested this algorithm on a case that was already solved, experimenting to see if his system could identify the Green River Killer who killed at least 71 women in Washington state during the 1980s and 1990s. According to Bloomberg, he reverse-engineered the algorithm, settling on four characteristics for a cluster analysis: geography, sex, age group, and method of killing. “It became clear that this thing was working,” Hargrove told Bloomberg. “In fact, it was working too well.” The algorithm went on to link over 100 unsolved murders of women in Phoenix and Los Angeles, many of which were attributed to several different people. 

While work still needs to be done, “this in itself serves as proof of concept that it may be possible for computer wizardry and data to be used to hunt down serial killers,” Tom Hale writes.