How Not to Get Away with Murder: The Serial Killer Detector

Each year, five thousand people get away with committing murder in the United States. And it’s likely they’ll strike again. Recognizing that a third of homicide cases goes unsolved, Thomas Hargrove is determined to develop a serial killer detector to find these murderers.

Since 2010, Hargrove, a homicide archivist, has been gathering information from municipal records on murders committed as far back as 1976. His catalogue contains information on 751,785 cases, making it the largest archive in the United States. He conceived the idea of using information to find serial killers while he was working as a journalist. Looking at the data provided by the FBI’s Supplementary Homicidal Report one day, he speculated whether statistical trends could help narrow down the searches. Hargrove told The New Yorker, “The first thing I thought was, I wonder if it’s possible to teach a computer to spot serial victims.”

He began writing code to sift through his catalogue. He started by telling the computer to search for statistical anomalies to vet out “ordinary murders” that arise due to love triangles, gang fights, and robberies. The serial killer detector then aggregates data based on method, geographical location, time, and the victim’s sex to highlight patterns.  This code forms the backbone of the Murder Accountability Project (MAP).

Developing the serial killer detector was difficult. Isaac Wolf said, “[Hargrove] would write some code, and it would run through what seemed like an endless collection of records. And we did not have expensive computer equipment, so it would run for days.” From these countless tests, Hargrove was able to narrow down his search parameters to sex, weapon, age, and location. There were certain patterns that emerged, such as women accounted for 70% of serial killer victims. Weapon was typically strangulation or bludgeoning. Significantly, Hargrove found that geographical location offered key insight into tracking down a serial killer. According to a New York City homicide detective, “Serial killers tend to stick to a killing field. They’re hunting for prey in a concentrated area, which can be defined and examined.” Murderers are less likely to act the further they are from the “hunting ground”.

Although not yet perfected, Hargrove’s algorithm has been useful in linking instances of murder to the possibility of a serial killer. For instance, Lake County in Indiana had fifteen cases of women who were strangled to death between 1980 and 2008. Studying the data, Hargrove suggested to the police of the city of Gary, which is located in Lake County, the possibility that these women were killed by the same person. The police rejected his report but four years later, the police in Hammond, a town near Gary, arrested a man named Darren Vann who was found to have been the perpetrator behind the Lake County killings.

There are still limitations to the serial killer detector. For one, it relies on data supplied by municipalities. Some cities are less capable of solving murders and documenting cases than others. The algorithm’s reliance on geographical data also makes it more difficult to find serial killers who travel farther than the typical hunting ground radius. Nevertheless, it’s a huge breakthrough.

Hargrove said, “Our primary purpose is to gather as many records as possible. It’s seductive how powerful these records are, though. Just through looking, you can spot serial killers. In various places over various years, you can see that something god-awful has happened.”

Are you experimenting with algorithms to solve unsolved mysteries like Hargrove? Did you know that your experiments, even the unsuccessful ones, could qualify for the R&D Tax Credit and you could receive up to 14% of your research expenses. To find out more, please contact a Swanson Reed R&D Specialist today or check out our free online eligibility test.

Swanson Reed regularly hosts free webinars and provides free IRS CE credits as well as CPE credits for CPA’s.  For more information please visit us at www.swansonreed.com/webinars or contact your usual Swanson Reed representative.

 

Recommended Posts