EEG-Based Emotion Recognition Using Deep Learning Network with Principal Component Based Covariate Shift Adaptation

By Suwicha Jirayucharoensak, Setha Pan-Ngum, Pasin Israsena. TheScientificWorldJournal, 2014

472 citations recorded by the source corpus.

Automatic emotion recognition is one of the most challenging tasks. To detect emotion from nonstationary EEG signals, a sophisticated learning algorithm that can represent high-level abstraction is required. This study proposes the utilization of a deep learning network (DLN) to discover unknown feature correlation between input signals that is crucial for the learning task. The DLN is implemented with a stacked autoencoder (SAE) using hierarchical feature learning approach. Input features of the network are power spectral densities of 32-channel EEG signals from 32 subjects. To alleviate overfitting problem, principal component analysis (PCA) is applied to extract the most important components of initial input features. Furthermore, covariate shift adaptation of the principal components is implemented to minimize the nonstationary effect of EEG signals. Experimental results show that the DLN is capable of classifying three different levels of valence and arousal with accuracy of 49.52% and 46.03%, respectively. Principal component based covariate shift adaptation enhances the respective classification accuracy by 5.55% and 6.53%. Moreover, DLN provides better performance compared to SVM and naive Bayes classifiers.

Subfields: Cognitive Neuroscience, Experimental and Cognitive Psychology, Signal Processing

View Paper: https://www.semanticscholar.org/paper/245013c50ce350dd225ab6238f374af9a62f5b89

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Experimental and Cognitive Psychology

In Signal Processing

Citations in the corpus

Deep Feature Learning for EEG Recordings (2015, background): “There has also been early work on emotion recognition from EEG using deep neural networks such as described by Jirayucharoensak et al. (2014) and Zheng et al. (2014).”
Applications of Deep Learning and Reinforcement Learning to Biological Data (2017, methodology): “used PCA extracted power spectral densities from each EEG channel, which were corrected by covariate shift adaptation to reduce non-stationarity, as features to stacked DA to detect emotion [160].”
Hierarchical Convolutional Neural Networks for EEG-Based Emotion Recognition (2017, unknown): “Their results indicate that SAE is better than SVM and naive Bayes classifiers [36].”
Deep learning for healthcare applications based on physiological signals: A review (2018, background): “, 2014 [118] Based Emotion Recognition Deep learning network DEAP Dataset [119] valence accuracy: 49.”
A Deep Learning Scheme for Motor Imagery Classification based on Restricted Boltzmann Machines (2017, methodology): “Jirayucharoensak et al [28] proposed an EEG-based emotion recognition method based on stacked autoencoders in combination with power spectral density and principle component analysis (PCA).”
Emotions Recognition Using EEG Signals: A Survey (2019, unknown): “[86] 2014 x x x x x x x x 57.”
Deep Learning in Mining Biological Data (2020, methodology): “A stacked DA was employed to detect emotion from EEG signals after extracting relevant features using PCA and reducing nonstationary effect using covariate shift adaptation [119].”
Deep learning in bioinformatics (2016, methodology): “[163] used DBN and SAE, respectively, for emotion classification.”
Real-Time Movie-Induced Discrete Emotion Recognition from EEG Signals (2018, methodology): “Then, we subtract the baseline power of each resting period from all of the corresponding segments of the extracted features, in such a way that the change in the power is based only on the current film clip and is not affected by the previous clips [41].”
A novel deep learning approach for classification of EEG motor imagery signals (2017, methodology): “A network with three autoencoders and two softmax layers was proposed in [26] for automatic emotion recognition from EEG signals.”