Feature Extraction and Selection for Emotion Recognition from EEG

By Robert Jenke, Angelika Peer, Martin Buss. IEEE Transactions on Affective Computing, 2014

967 citations recorded by the source corpus.

Emotion recognition from EEG signals allows the direct assessment of the “inner” state of a user, which is considered an important factor in human-machine-interaction. Many methods for feature extraction have been studied and the selection of both appropriate features and electrode locations is usually based on neuro-scientific findings. Their suitability for emotion recognition, however, has been tested using a small amount of distinct feature sets and on different, usually small data sets. A major limitation is that no systematic comparison of features exists. Therefore, we review feature extraction methods for emotion recognition from EEG based on 33 studies. An experiment is conducted comparing these features using machine learning techniques for feature selection on a self recorded data set. Results are presented with respect to performance of different feature selection methods, usage of selected feature types, and selection of electrode locations. Features selected by multivariate methods slightly outperform univariate methods. Advanced feature extraction techniques are found to have advantages over commonly used spectral power bands. Results also suggest preference to locations over parietal and centro-parietal lobes.

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

View Paper: https://www.semanticscholar.org/paper/5c490be1858af6a6b55e238f0f4a6b2b85613095

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

EEG-Based Emotion Recognition via Channel-Wise Attention and Self Attention (2023, methodology): “In the past decade, the relationship between emotion and EEG signals has been studied extensively [9], [10], [11].”
A Novel Bi-Hemispheric Discrepancy Model for EEG Emotion Recognition (2019, methodology): “[7] evaluated all the existing features by using machine learning techniques on a”
An Efficient LSTM Network for Emotion Recognition From Multichannel EEG Signals (2020, background): “Feature extraction plays an important role in EEG-based emotion recognition [31].”
Physiological Signals-based Emotion Recognition via High-order Correlation Learning (2019, background): “…into the following three modes according to the type of a channel [26]: (a) visual feedback on emotional stimuli, such as body language, postures and facial expressions, (b) written or the verbal expressions, and (c) physiological signals of the change of physiological features [16, 41].”
Identifying Stable Patterns over Time for Emotion Recognition from EEG (2016, background): “The findings of these neural patterns are consistent with previous emotion studies [14], [17], [42], [59], [61].”
EmotionMeter: A Multimodal Framework for Recognizing Human Emotions (2019, background): “science support the hypothesis that brain activity is associated with emotions [2], [12], [13] and that EEG allows for the direct assessment of the “inner” states of users [27].”
A Bi-Hemisphere Domain Adversarial Neural Network Model for EEG Emotion Recognition (2021, background): “Jenke et al. [8] provided a comprehensive survey on the EEG feature extraction approaches.”
EEG-Based Emotion Recognition: A State-of-the-Art Review of Current Trends and Opportunities (2020, background): “Works describing that EEG contains the most comprehensive features such as the power spectral bands can be utilized for basic emotion classiﬁcations [50].”
EEG-Based Emotion Recognition Using Regularized Graph Neural Networks (2019, background): “EEG features can be broadly divided into single-channel features and multi-channel ones [25].”
EEG Based Emotion Recognition: A Tutorial and Review (2022, background): “[126] extracted time-frequency-domain features as well as channel combination features from multichannel EEGs.”