Move over Sherlock: AI detective Valcri solves crimes using deductions made from MILLIONS of pieces data

Move over Sherlock: AI detective Valcri solves crimes using deductions made from MILLIONS of pieces data

Move over Sherlock: AI detective Valcri solves crimes using deductions made from MILLIONS of pieces data

0 comments 📅15 May 2017, 00:06

Sherlock Holmes is famed for his ability to connect the dots between seemingly unrelated pieces of information.

Now scientists have developed an AI system that can match the super-sleuths crime solving capabilities.

Valcri scans millions of pieces of data to suggest how and why a crime may have been committed, freeing investigators to work on building their case.

Researchers have created Valcri, an AI super-sleuth that can scan millions of pieces of data to suggest how and why a crime may have been committed, which analysts can access in more detail via interactive touchscreens (stock image)

The software began life at the University of Middlesex in London, who received a €13 million (£10 million / $17 million) investment from the European Union to fund the project in 2014.

Valcri, which stands for Visual Analytics for sense-making in Criminal Intelligence Analysis, was originally conceived as a way to provide early warnings of impending criminal activity, including online fraud and global terrorism.

But the system can also be implemented to solve the more routine work of law enforcement agencies.

West Midlands Police has been testing Valcri with around 6.5 million anonymised records gathered over three years, as well as a force in Antwerp, Belgium.

And Valcri’s creators hope to implement the system in real-time in the near future, according to reports in New Scientist.

Speaking to the website William Wong, who leads the project at Middlesex University, said: ‘Everyone thinks policing is about connecting the dots, but that’s the easy bit.

‘The hard part is working out which dots need to be connected.’

Valcri works by studying police reports, including text, images, and videos, to suggest links that may be worth further investigation.

This data is then presented to crime analysts via touchscreens, which allow them to examine each piece of evidence in closer detail.

Crime analysts can spend hours trawling through records, but Valcri allows them to access relevant information at the touch of a button.

And by using the latest in machine learning and facial recognition technology, the system can improve its results based on input from the analysts.

Neesha Kodagoda, an expert in human computer interactions from the university, added: ‘An experienced analyst needs 73 individual searches to gather all of this information, before manually putting it into an easily digestible form.

‘Valcri can do this with a single click.’

Source: http://www.dailymail.co.uk