EMOTION RECOGNITION FROM ELECTROENCEPHALOGRAPHIC SIGNALS USING ARTIFICIAL INTELLIGENCE

This thesis examines the application of artificial intelligence in emotion recognition using EEG signals, with a focus on a subjective model, specifically, one individual. The choice of a single-subject approach is due to the substantial inter-subject variability inherent in EEG data. Such variability often requires the collection of large amounts of data to allow generalization across individuals, which is currently impractical given the difficulties associated with collecting EEG data from multiple subjects. By focusing on the EEG signals of a single individual, this research aims to develop a powerful artificial intelligence model that can accurately predict the emotional state of a single individual.

The methodology of this study involves the systematic collection of EEG data under controlled experimental conditions for a single individual, designed to elicit different emotional responses. The collected EEG data is then pre-processed to remove noise and artifacts, ensuring the quality and reliability of the signals. Machine learning and deep learning algorithms are used to classify emotional states presented in EEG data.

To summarize, this thesis presents an approach to emotion recognition using EEG signals and artificial intelligence, demonstrating the potential for developing accurate, reliable, and personalized emotion recognition systems. Refined research on one subject is a good starting point for future generalizations to many people; from this point, it is already possible to apply it to various human-computer interaction modules.