قالب وردپرس درنا توس
Home / Science / Text-based lie detector can help cops filter out fake Robbery reports

Text-based lie detector can help cops filter out fake Robbery reports



Computer-assisted spelling and grammar checking has been around for decades, but what if software could also function as a polygraph for text-based documents? That may sound crazy, but it's an issue that researchers from Cardiff University and Charles III University have been working on in Madrid – and it's obviously paying off. What the computer scientists have developed is a tool that can help the police find false statements about robberies. This is a major problem in certain parts of the world and can lead to a number of negative consequences – including false insurance claims, wasted police time and possibly even illegal arrests. The consequences of a false police report can include heavy fines and even jail sentences.

VeriPol, the new text analysis tool, uses machine learning techniques to examine information such as the nature of stolen articles and descriptions by perpetrators. According to its creators, it can then identify false police reports with an accuracy of more than 80 percent. Frequent repetitions of false reports include a greater focus on the stolen property than the incident, lack of exact details, limited details of the attacker, and lack of witnesses or hard evidence.

"The main objective is to create a tool that will be 'helpful for early detection of these fake police reports so that it can be an adjunct to police knowledge," said Dr. Jose Camacho Collados, a Cardiff research associate University, towards Digital Trends. "In particular, in [our research] we are trying to specifically investigate whether such a tool can be automatically learned from the textual content of police reports and applied to new police reports. We used classical machine learning approaches and recorded police reports that were known to be right or wrong. The main knowledge was that people with similar patterns are lying and that a classical machine learning approach can recognize such patterns and can be effectively applied to this problem with a high degree of accuracy. "

VeriPol has been used in several pilot studies that have shown how it can be successfully used to filter out false predator reports." We have now integrated this system into the Spanish police architecture, along with a comprehensive training program, and it will be launched in As a side effect ̵

1; and perhaps more importantly – we expect the awareness and communication of this problem will prevent people from writing fake police reports. "











Source link