Automatic Recognition of Human Emotions from Electroencephalography Signals System

Abstract

Recognizing someone’s emotional state could be important in our daily life affairs.  Some people’s emotional status can affect their performance in work either positively or negatively and they might react to the situation according to their status. Recognizing a person’s emotional state can be carried by body language and facial expressions. However, people can pretend to express an opposite emotional state and hide the actual one. Using an Artificial Intelligence (AI) system of electroencephalography (EEG) human brain signals can recognize the actual person’s emotional status and feelings. The proposed model stages would use a free dataset of DEAP Learning available on the internet for human EEG signals, filter the data in the pre-processing stage and specify the training and testing data, use the LSTM algorithm in the feature extraction stage to create the final model, verify the system using the testing to identify the accuracy system level. The data obtained by the analysis of EEG signals can be used in many fields like education, science, hospitals, security, and research. The design of Automatic Recognition of Human Emotions from an Electroencephalography Signals system can be considered sustainable and required for future (AI) systems development which is a growth sector around the world.