A few years ago I spoke with Babak Akghar, the lead researcher in the ATHENA project, which aimed to utilize technology in crime fighting. Since then, a lot has happened in the industry, with virtual reality used to recreate crime scenes, the use of machine learning to better manage DNA samples, and various attempts at predictive policing, both in public and inside stores.
One of the more interesting innovations has been the use of facial recognition technology. Schipol airport have been dabbling with the technology to smooth passport checks, whilst in China, the police have been testing the use of facial recognition technology on police cars.
Now, the technology is coming to the UK, with police using algorithms to help their analyses. The system, called VALCRI (Visual Analytics for sense making in CRiminal Intelligence analysis), has been developed by a European consortium headed by Middlesex University. It aims to ensure policing is more intelligence led, both now and in the future.
The system analyzes anonymous criminal records, whilst also pulling in data from a range of other sources to provide the analysts with insight into both social and legal factors that may influence the work of detectives. It’s a process that is done ordinarily today but requires a huge amount of time and manpower, with a high likelihood of errors or biases influencing the work.
The team believe VALCRI overcomes that and provides invaluable input into the crime fighting process, without attempting to fight the crimes itself. The AI backend to the system learns as new data is added. Analysts are presented with a graphical interface to make deriving insights as easy as possible.
Whilst its primary aim is to support ongoing investigations, the team also believe that VALCRI can provide a predictive capability to police forces courtesy of the patterns it can detect in crime data. The data will also be made available for criminal researchers working in the field to help inform better policy.
The project is currently being piloted by both Brussels police force and West Midlands Police Force, and it will be interesting to see how it goes. I wrote last year about a report that revealed less than stellar results from a predictive policing project in Chicago.
The project, which aimed to reduce gun crimes, used models to trawl through arrest data and provide a list of people that were believed to be at high risk of gun crime. The hope was that this data could then be used to keep them from harm.
The report highlighting two main issues with the approach. Firstly, the team struggled to get officers to actually use the data, with around 2/3 of cases seeing the list produced by the model ignored.
This was often because officers felt ill equipped to deal with the information available to them, with no apparent direction given as to how to manage those individuals contained on the list. Rather than being empowered by the data therefore, the officers often felt confused by it.
Hopefully VALCRI will do significantly better.