MULTI-FEATURE FUSION EMOTION RECOGNITION BASED ON RESTING EEG

• DOI: 10.1142/s0219519422400024
• Publication Date: 2022-03-31T09:48:47Z
• Authors: JUN-AN ZHANG; LIPING GU; YONGQIANG CHEN; GENG ZHU; LANG OU; LIYAN WANG; XIAOOU LI; LICHANG ZHONG
• Keywords: Normalization; Feature (linguistics)

An important task of brain–computer interface (BCI) research is to read or decode mental content from a large number electroencephalographic (EEG) signals, which an exciting, attractive, and challenging goal. In this paper, resting EEG data are collected spontaneous EEG, 16 random features such as correlation coefficient, covariance, brainpower spectrum extracted reference vectors. the subsequent emotion recognition experiment, same feature information separated signal, translation normalization processing carried out based on resting-state features. Finally, with machine learning methods [Formula: see text]-means clustering multi-feature fusion, positive, negative, neutral emotional characteristic parameters were correctly separated. group 12 subjects, correct rate visual evoked emotions reached 83.9%, was better than literature mentioned in paper. Another highlight method that it can quickly, accurately, efficiently select best matching least resource consumption multiple potential acquisition points. Further analysis comparison characteristics find relationship between specific stimuli corresponding signals.