When dealing with patients with psychological or emotional symptoms, medical practitioners are often faced with the problem of objectively recognizing their patients' emotional state. In this paper, we approach this problem using a computer program that automatically extracts emotions from EEG signals. We extend the finding of Koelstra et. al [IEEE trans. affective comput., vol. 3, no. 1, pp. 18-31, 2012] using the same dataset (i.e. the DEAP: dataset for emotion analysis using electroencephalogram, physiological and video signals), where we observed that the accuracy can be further improved using wavelet features extracted from shorter time segments. More precisely, we achieved accuracy of 65% for both valence and arousal using the wavelet entropy of 3 to 12 seconds signal segments. This improvement in accuracy entails an important discovery that information on emotions contained in the EEG signal may be better described in term of wavelets and in shorter time segments.
Objectively recognizing emotions is a particularly important task to ensure that patients with emotional symptoms are given the appropriate treatments. The aim of this study was to develop an emotion recognition system using Electroencephalogram (EEG) signals to identify four emotions including happy, sad, angry, and relaxed. We approached this objective by firstly investigating the relevant EEG frequency band followed by deciding the appropriate feature extraction method. Two features were considered namely: 1. Wavelet Energy, and 2. Wavelet Entropy. EEG Channels reduction was then implemented to reduce the complexity of the features. The ground truth emotional states of each subject were inferred using Russel's circumplex model of emotion, that is, by mapping the subjectively reported degrees of valence (pleasure) and arousal to the appropriate emotions - for example, an emotion with high valence and high arousal is equivalent to a `happy' emotional state, while low valence and low arousal is equivalent to a `sad' emotional state. The Support Vector Machine (SVM) classifier was then used for mapping each feature vector into corresponding discrete emotions. The results presented in this study indicated thatWavelet features extracted from alpha, beta and gamma bands seem to provide the necessary information for describing the aforementioned emotions. Using the DEAP (Dataset for Emotion Analysis using electroencephalogram, Physiological and Video Signals), our proposed method achieved an average sensitivity and specificity of 77.4% ± 14.1% and 69.1% ± 12.8%, respectively.
Emotion classification is one of the state-of-the-art topics in biomedical signal research, and yet a significant portion remains unknown. This paper offers a novel approach with a combined classifier to recognise human emotion states based on electroencephalogram (EEG) signal. The objective is to achieve high accuracy using the combined classifier designed, which categorises the extracted features calculated from time domain features and Discrete Wavelet Transform (DWT). Two innovative designs are involved in this project: a novel variable is established as a new feature and a combined SVM and HMM classifier is developed. The result shows that the joined features raise the accuracy by 5% on valence axis and 1.5% on arousal axis. The combined classifier can improve the accuracy by 3% comparing with SVM classifier. One of the important applications for high accuracy emotion classification system is offering a powerful tool for psychologists to diagnose emotion related mental diseases and the system developed in this project has the potential to serve such purpose.
Due to the high dimensional, non-stationary and nonlinear properties of electroencephalogram (EEG), a significant portion of research on EEG analysis remains unknown. In this paper, a novel approach to EEG-based human emotion study is presented using Big Data methods with a hybrid classifier. An EEG dataset is firstly compressed using Compressed Sensing (CS), then, wavelet transform features are extracted, and a hybrid Support Vector Machine (SVM) and Fuzzy Cognitive Map (FCM) classifier is designed. The compressed data is only one-fourth of the original size, and the hybrid classifier has the average accuracy by 73.32%. Comparing to a single SVM classifier, the average accuracy is improved by 3.23%. These outcomes show that psychological signal can be compressed without the sparsity identity. The stable and high accuracy classification system demonstrates that EEG signal can detect human emotion, and the findings further prove the existence of the interrelationship between various regions of the brain.
Abstract-Recognizing emotion from EEG signals is a complicated task that requires complex features and a substantial number of EEG channels. Simple algorithms to analyse the feature and reduce the EEG channel number will give an indispensable advantages. Therefore, this study explores a combination of wavelet entropy and average wavelet coefficient (WEAVE) as a potential EEG-emotion feature to classify valence and arousal emotions with the advantage of the ability to identify the occurrence of a pattern while at the same time identify the shape of a pattern in EEG emotion signal. The complexity of the feature was reduced using the Normalized Mutual Information (NMI) method to obtain a reduced number of channels. Classification with the WEAVE feature achieved 76.8% accuracy for valence and 74.3% for arousal emotion, respectively. The analysis with NMI shows that the WEAVE feature has linear characteristics and offers possibilities to reduce the EEG channels to a certain number. Further analysis also reveals that detection of valence emotion with reduced EEG channels has a different combination of EEG channels compared to arousal emotion.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.