Gender Based Emotion Classification Using M-GRU Based on Brain EEG Signal

The need for computer programs that can identify a user's present emotional state is increasing all the time. Research on identifying emotion from the face and voice has previously been done in an attempt to mimic human communication. Humans can discern emotions with 70-98 percent accuracy from these signals, and computers are already fairly good at classifying facial expressions (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$80-90$</tex> percent). These programs train the brain's response by using information about the test subject's mental or emotional state. Computational models called neural networks are modeled after the structure and operation of the human brain. They have artificial neurons that are interconnected, arranged in layers, and are able to learn and make decisions depending on input data. For this study, one male and one female volunteer were chosen. Our findings suggest that brain EEG waves can be used to detect emotions. Cortical activity is reconstructed with high resolution and spatiotemporal precision using the new multi-mode EEG/MRI integrated approach. A new application of BCI proposed in this research is the prediction of emotion based on a person's gender using EEG analysis using Modified Gated Recurrent Unit (M-GRU). The accuracy of emotion recognition based on gender classification difficulties was <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 5. 3 1 2}$</tex> percent.