Deception
From Distributed Information and Intelligence Analysis Group
Deception detection plays an important role in safely and reliably using expert systems in areas such as battlefield, intelligence community and politics. The benevolence assumption people have based their implementations of expert (human and/or synthetic) systems on is rarely valid in the real world. Unfortunately, deception detection is extremely challenging. The average detection rate by humans alone is only above chance and human detection skill has shown to be difficult to improve even with training. Traditional deception detections focus on human’s non-verbal cues which might not be available or precisely captured, while we detect deception from the reasoning processes of experts.
