The latest lie detector software seems to copy real court cases data in its repertoire.
Machine learning is a brand new world unto itself. It has allowed computers to find out a person’s age, count the calories in a food item and even take over one’s occupation.
Don't Miss: Sam's Club Black Friday 2016 Details
Scientists at the University of Michigan have taken to using this thing called machine learning to detect lies. Testimonies from real life court cases were employed in the sequences. The machines were used to decipher whether a person was telling a lie or not.
The machine learning used both the words and the gesticulations the person made while lying. The surprising thing was that the computer software was accurate 75% of the time as far as pinpointing fibs was concerned.
But human beings were accurate only 50% of the time. This is quite an achievement for artificial intelligence. In order to spot the Pinocchios among the lot, the machine honed in on various behaviors and styles of speaking.
Eye contact, confident tone of voice and gestures were among the key markers. Since people normally use a high degree of linguistic skill, they also happen to be poor detectors of lies. That is because most of the truth resides in body language.
"In laboratory experiments, it's difficult to create a setting that motivates people to truly lie. The stakes are not high enough," said Rada Mihalcea, professor of computer science and engineering who leads the project with Mihai Burzo, assistant professor of mechanical engineering at UM-Flint.
"We can offer a reward if people can lie well—pay them to convince another person that something false is true. But in the real world there is true motivation to deceive."
The focus was not on how many times a liar referred to himself or herself. Nor was it on how many times he or she looked up. Small details included the beating of his or her heart, respiratory flow and body temperature flux were crucial.
Thermal imaging was used to record these small details. Computers also may measure various gestures. This leaves not even the smallest clues to pure chance. A lie can be caught in a matter of seconds.
"People are poor lie detectors," Mihalcea said. "This isn't the kind of task we're naturally good at. There are clues that humans give naturally when they are being deceptive, but we're not paying close enough attention to pick them up. We're not counting how many times a person says 'I' or looks up. We're focusing on a higher level of communication."
What was supposed to be a secret would no more be a secret at all. Thus by using court cases that have high stakes involved, researchers are able to build the best lie detectors on the planet.
These don’t work in the same manner as a polygraph which is an old hat by now. Lying individuals also tended to fidget and use their hands more often. This was a vital clue as to whether a person was telling the truth or just lying.
By the way, the subject does not need to be touched to know whether he or she is lying in the new procedures. The system has quite some real world applications. Security agents could use this to detect little white lies. Via the integration of body language and vital signs, lies could be caught in a jiffy.
A paper on the findings titled "Deception Detection using Real-life Trial Data" was presented at the International Conference on Multimodal Interaction and is published in the 2015 conference proceedings.