Researchers at Columbia University and the City University of New York (CUNY) are developing computational methods to detect deception in English, Mandarin Chinese, and Arabic speakers.
The researchers, led by Columbia professor Julia Hirschberg, hope that the work can lead to the development of lie-detection technology that is more accurate than the polygraph test.
As part of the study, 32 subjects were asked to complete random, unrelated tasks such as tying knots, stacking quarters on their elbows, hopping on one foot, and singing, all of which were tasks they were told the U.S.'s top 25 entrepreneurs also were tested on. All of the participants were told they had failed to perform as well as the entrepreneurs, but that they must try to convince an interviewer that they had performed just as well, pressing a hidden pedal when they lied and another when they told the truth.
Using the data, the researchers built classifiers using machine learning that were about 70 percent accurate in determining truth from lies, while human judges were only about 58 percent accurate.
From Columbia University
View Full Article
No entries found