Rigel and the Rapist
How an algorithm ended an 11-year crime spree.
Infographic by Stephen Rountree
A computer will never replace a human investigator, but it can help focus the search. In the South Side Rapist case, Kim Rossmo’s Rigel program analyzed crime scene locations. His work narrowed the suspect list, which ultimately led to an arrest. Here’s how it works.
1) Mark the Crime Sites
Rigel plots the crime scene locations on a map.
2) Mold the Profile
Offenders typically strike in familiar places, moderately close to their home or work, yet leave a “buffer zone” to protect their anonymity. Rigel expresses these tendencies with a probability function. The function decays with distance to reflect the reduced chance the offender lives far from a crime scene, and dips in the middle since offenders rarely live adjacent to the site.
3) Copy and Paste
Rigel applies this function to the location of every crime. The profile grows more
accurate as sites are added.
4) Sum the Data
By combining all these
probability functions, Rigel creates a peaked map that highlights the area where the offender is most likely to live.
5) Overlay the Suspects
Investigators compare a list of suspects’ homes and places of work against the probability map, and focus the search on suspects based near the peak of the curve. In this case, the rapist lived in the center of Rossmo’s hot zone. He was positively identified using DNA and is now serving 3 consecutive life sentences.