Source: Cardiff UniversityComputer science experts from Cardiff University and Charles III University of Madrid have created a tool that helps law enforcement determine whether the contents of a written police statement are fake.

The computer tool relies on a combination of machine learning techniques and automatic text analysis to make such determinations and has successfully spotted fake robbery reports with a rate of accuracy over 80%, according to the experts.

Currently in use all over Spain, the tool, called VeriPol, helps officers determine whether an investigation needs to be expanded.

VeriPol is designed specifically to be used on robbery reports. In some parts of the world, false reports are commonplace, contaminating police databases and wasting vast amounts of resources.

As such, the team designed VeriPol to locate patterns common with false reports, such as information concerning the items reported stolen, descriptions of the alleged perpetrator(s) and the finer details of the incidents reported.

The tool, according to researchers, might save law enforcement time and effort, working in combination with traditional investigative methods. Likewise, VeriPol might also discourage those filing the false reports altogether.

To train the system, VeriPol was fed data from known false police reports so that it could “learn” the patterns.

Once trained, VeriPol was tested on over 1,000 police reports from the Spanish National Police and was, according to the team, “extremely effective in discriminating between false and true reports.”

Among the themes commonly appearing in false robbery reports were abbreviated statements offering more information on the items stolen than on the incident itself; fewer details about the incident and about the attacker; and no witnesses or substantial evidence.

"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," co-author of the study Dr. Jose Camacho-Collados, from Cardiff University's School of Computer Science and Informatics, said.

"Our study has given us a fascinating insight into how people lie to the police, and a tool that can be used to deter people from doing so in the future," continued Dr. Camacho-Collados.

"Police officers across Spain are now using VeriPol and integrating it into their working practices. Ultimately we hope that by showing that automatic detection is possible it will deter people from lying to the police in the first instance," Dr. Camacho-Collados added.