User:AlexandrKalaur

Brainternet is a project of students at the University of the Witwatersrand, a neurocomputer interface that reads data on the activity of the human brain for the subsequent use of this information for machine learning.

As a tool for data collection, scientists used the 14-channel wireless electroencephalograph EMOTIV EPOC, the average cost of which is about 800 US dollars (2017). Sensors of electroencephalograph collect electrical impulses of the cerebral cortex in 14 areas of the frontal, temporal, parietal and occipital lobes. After the collection, the data is processed by the method of reference vectors, one of the types of training with the teacher. Further, on the basis of these data, the algorithm links the work of the brain - invisible to the person's eye - and the specific physical operations of the person in the external environment. As a result, Brainternet is able to determine some actions of an individual, for example, based on the data received, the algorithm can determine whether a person is standing or walking.

The project is of great importance both for improving machine learning and neurocomputer interfaces, and for science in general.

"Today there is not enough visual (for the layman) data on how the human brain processes information. Brainternet is designed to make it easier for the user to understand the principles of the work of his own brain and brains of other people."