Scientists have come up with a way to tell if someone has lied on a police statement related to robberies using a computer tool that analyses text.
VeriPol, developed by computer science experts from Cardiff University and Charles III University of Madrid, is now being used by police officers across Spain, after a trial detected false claims in robbery reports with over 80% accuracy.
The system aims to help ease the work of officers using traditional investigative techniques, as well as saving precious time.
Using automatic text analysis and advanced machine learning, the technology can spot common patterns associated with fake robbery statements given to police, such as the types of items said to have been stolen, details of incidents and descriptions of a perpetrator.
“As an example, our model began to identify false statements where it was reported that incidents happened from behind or where the aggressors were wearing helmets,” said Dr Jose Camacho-Collados, from Cardiff University’s School of Computer Science and Informatics.
“Similarly, other clear indicators of falsehood were descriptions of the type of objects stolen. References to iPhones and Samsung were associated with false claims, whereas bicycles and necklaces were correlated with true reports.”
During a trial of the technology with Spanish police around Murcia and Malaga in June 2017, a total of 64 false robbery reports were detected and closed within one week, over three times higher than the average number of false reports detected by police in the same month between 2008 and 2016.
Ultimately, police authorities hope that the new technique will deter people from lying in their statements.
In the UK, making a false report can lead to a fine of up to £80, a conviction for wasting police time and even a prison sentence for perverting the course of justice.