Multichannel Optimization With Hybrid Spectral- Entropy Markers for Gender Identification Enhancement of Emotional-Based EEGs

Investigating gender differences based on emotional changes supports automatic interpretation of human intentions and preferences. This allows emotion applications to respond better requirements customize interactions affective responses. The electroencephalogram (EEG) is a tool that potentially can be used detect differences. main purpose this paper twofold. Firstly, it aims use both linear nonlinear features EEG signals identify influences behavior. Secondly, develop an recognition model by employing optimization algorithms the most effective channels for identification from emotional-based signals. EEGs thirty healthy students University Vienna were recorded while they watched four short video clips depicting emotions anger, happiness, sadness neutral. In study, wavelet transform (WT) de-noising technique, spectral mean frequency ( meanF) multiscale fuzzy entropy MFE) used. individual performance these attributes was statistically examined using analysis variance (ANOVA) represent behavior in brain-emotion females males. Then, two fused into set hybrid spectral-entropy SEA). Consequently, including binary gravitation search algorithm (BGSA) particle swarm (BPSO), employed optimal classification. Finally, k-nearest neighbors kNN) classification technique dataset. results show are remarkable neuromarkers investigating gender-based states. Moreover, significant enhancement overall accuracy achieved BGSA with proposed SEA when compared features. Therefore, methods improving process