Researchers have been successful in making it possible that computers may learn on their own through the use of machine learning. This opens the door to a plethora of new applications, and only recently, researchers from Japan working at the University of Tokyo drew up ideas for developing a lie detector that would use both artificial intelligence and machine learning. An advancement of this kind can be beneficial since it is precisely the kind that can potentially find applications in various domains, one of which is preventing criminal activity.
After all that has been stated and now that it is out of the way, it is crucial to understand that this deception detection system can examine an individual’s face to determine whether or not they are telling the truth about what they are saying. To acquire data from four male graduate students, the participants requested that they purposely lie about visuals that they were being shown.
The researchers used a camera to record the facial expressions the participants produced while they were lying, following the instructions given to them. When the session was over, the participants were questioned about whether or not they had lied during it and whether or not they had told the truth throughout it. This was done after everything had been thought through and taken into consideration.
The method of machine learning known as Random Forest was utilized in developing the deception detection model. This model accumulated data consisting of things like pulse measures and facial expressions. This data collection will likely be used to make the AI considerably more accurate than it would have been otherwise.
It was predicted that the model had an accuracy of between 75% and 80%, and it had the potential to become a very valuable instrument in all areas. Even though this is the case, the dataset is still rather restricted. To determine whether or not the results can still be replicated, the study will need to be carried out on a much larger scale. It will be fascinating to watch if or not additional tests of this sort are carried out in the future since they will be vital for establishing the overall efficacy of this deception test and others like it that are currently being created. It will be interesting to observe if further tests of this variety are carried out.
