Computer systems have feelings as well: New analysis shows artificial intelligence can teach technologies to recognize feelings with 98% accuracy

The rise of artificial intelligence is 1 of the most influential and talked about technological developments in the globe. Its quickly expanding capabilities have incorporated it into daily life, and now it sits in our living rooms and, some say, threatens our jobs.

Despite the fact that AI enables machines to operate with a degree of human-like intelligence, the 1 point humans have normally had more than machines is the potential to show feelings in response to the scenarios they obtain themselves in. But what if AI could be applied to allow machine and automatic emotion recognition technologies?

New analysis from Brunel University in London and Iran’s Bonab University and Islamic Azad University applied signals from EEG – a test that measures the brain’s electrical activity – and from artificial intelligence to create a laptop model for automatic emotion recognition to classify feelings with larger accuracy of 98%.

By focusing on information and instruction algorithms, computer systems can be taught to course of action information in the very same way that the human brain can. This branch of artificial intelligence and laptop science is known as machine mastering, exactly where computer systems are taught to imitate the way humans understand.

Dr Sebelan Danishvar, a analysis fellow at Brunel, stated: “The generative adversarial network, identified as GAN, is a essential algorithm applied in machine mastering that enables computer systems to mimic how the human brain performs. A person’s emotional state can be detected working with physiological indicators such as EEG. Considering that EEG signals originate straight from the central nervous method, they have a robust association with distinct feelings.

“Using GANs, computer systems understand how to carry out tasks soon after seeing examples and instruction information. They can then develop new information, which enables them to progressively strengthen accuracy.”

A new study, published in the journal Electronicsmusic stimulated the feelings of 11 volunteers, all in between the ages of 18 and 32.

The participants have been instructed to abstain from alcohol, drugs, caffeine and power drinks for 48 hours just before the experiment, and none of them had depressive issues.

For the duration of the study, all volunteers have been provided ten pieces of music to listen to via headphones. Delighted music was applied to evoke good feelings, and sad music was applied to evoke adverse feelings.

Although listening to music, participants have been connected to an EEG brain device, and the EEG signals have been applied to recognize their feelings.

In preparation for the study, the researchers made a GAN algorithm, working with an current database of EEG signals. The database contained information on feelings induced by musical stimulation, and this was applied as a model of them against genuine EEG signals.

As anticipated, music evoked good and adverse feelings, according to the music played, and the final results showed that there was a higher similarity in between the genuine EEG signals and the signals modeled by the GAN algorithm. This indicates that the GAN was efficient in creating EEG information.

Dr. Danishvar stated, “The final results show that the proposed strategy is 98.two% correct in distinguishing in between good and adverse feelings.” Compared with preceding research, the proposed model performed nicely and can be applied in future brain-laptop interface applications. This consists of the robot’s potential to discern human emotional states and interact with folks accordingly.

&#13

“For instance, robotic devices can be applied in hospitals to cheer sufferers up just before important operations and prepare them psychologically.”

&#13

“Future analysis should really discover more emotional responses in our GAN, such as anger and disgust, to make the model and its applications even much more valuable.”

Reported by:

Nadine Palmer, Media Relations

+44 ()1895 267090
nadine.palmer@brunel.ac.uk