EEG-Based Emotion Recognition Using Deep Learning and M3GP

This paper presents the proposal of a method to recognize emotional states through EEG analysis. The novelty this work lies in its feature improvement strategy, based on multiclass genetic programming with multidimensional populations (M3GP), which builds features by implementing an evolutionary technique that selects, combines, deletes, and constructs most suitable ease classification process learning method. In way, problem data can be mapped into more favorable search space best defines each class. After M3GP, results showed increment 14.76% recognition rate without changing any settings tests were performed biometric dataset (BED), designed evoke emotions record cerebral cortex’s electrical response; implements low cost device collect signals, allowing greater viability for application results. proposed methodology achieves mean 92.1%, simplifies management increasing separability spectral features.