EEG person identification using Facenet, LSTM-RNN and SVM

Electroencephalography (EEG) is the recording of electrical activity occurring in the brain, it represents the whole neural signal flow of the brain giving plenty of complex, yet abstract information. As many papers validated the individual identification using EEG Features for security purposes, we strongly believe that human EEG contains a fingerprint pattern that may identify each subject apart. In this paper, we have created a new Long Short-Term Memory (LSTM) based neural network that has learned pattern recognition from Google's, facial identification neural network, Facenet and forwarded by a Support Vector Machine (SVM) for final clustering. Our findings have shown, a pattern recognition with a loss of 0.137 and a final clustering accuracy of 97.84%.