Criminologists from the University of Cambridge are using 12 months’ worth of knife crime data to help make predictions about future fatal stabbing events in London neighborhoods.

In collaboration with a Metropolitan police detective, the criminologists determined that knife assaults that had taken place in a 12-month period potentially corresponded with an increased risk of fatal knife crimes in the same neighborhood the following year.

To make this determination, DCI John J. Massey of the Met’s Homicide Command manually sifted through thousands of knife crime records, selecting and geocoding events where victims were stabbed but survived during financial year 2016/2017. The team discovered that more than 3,500 knife assaults had occurred during that period of time, loosely translating to a ratio of 66 non-fatal stabbings for each knife homicide during that time. Those assaults were then coded to specific London neighborhoods and those locations were then measured against locations where 97 homicides took place during the following financial year's data (2017/2018).

The team of criminologists determined that of the 41 different London neighborhoods that had experienced six or more non-fatal knife assaults and injuries during the 2016/2017 financial year, 15% went on to experience a homicide during the 2017/2018 financial year. In other words, researchers explained that such neighborhoods are 15 times more likely to experience a knife homicide in the following year than neighborhoods in London who saw no knife assaults the year before.

The team believes that when developed further, the data could potentially result in technology that sends out real-time daily alerts to law enforcement concerning homicide risks, potentially better allocating resources to higher risk areas.

“If assault data forecasts that a neighbourhood is more likely to experience knife homicide, police commanders might consider everything from closer monitoring of school exclusions to localised use of stop-and-search," said study co-author Professor Lawrence Sherman from the University of Cambridge.

"Police IT is in urgent need of refinement. Instead of just keeping case records for legal uses, the systems should be designed to detect crime patterns for prioritising targets. We need to transform IT from electronic filing cabinets into a daily crime forecasting tool," he added.

The research is published in the Cambridge Journal of Evidence-Based Policing.