Open AccessAn Emotion Classification Method Based on Energy Entropy of Principal Component
Author(s) -
Hao Li,
Xia Mao,
Lijiang Chen
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1487/1/012002
Subject(s) - principal component analysis , pattern recognition (psychology) , artificial intelligence , support vector machine , dimensionality reduction , computer science , entropy (arrow of time) , electroencephalography , wavelet , curse of dimensionality , feature extraction , energy (signal processing) , classifier (uml) , speech recognition , mathematics , statistics , psychology , physics , quantum mechanics , psychiatry
Emotional recognition based on electroencephalogram (EEG) has attracted more and more attention, and various methods emerge in an endless stream. An emotion classification method based on energy entropy of principal component (PCEE) is proposed in this paper. EEG data are divided into five rhythms (δ, θ, α, β and γ) by wavelet decomposition and reconstruction (WDR). Each rhythm signal uses principal component analysis (PCA) to perform dimensionality reduction on the channels (electrodes). The energy entropies of the principal components that meet the requirements are used as the classification feature. Results show that the classification accuracy can reach 87.61% by using the support vector machine (SVM) classifier.