Emotional state classification from EEG data using machine learning approach

WangXiao-Wei,NieDan,LuBao-Liang

IF: 6

2014-01-01

Neurocomputing

Abstract:Recently, emotion classification from EEG data has attracted much attention with the rapid development of dry electrode techniques, machine learning algorithms, and various real-world applications ...

What problem does this paper attempt to address?

Emotional State Classification from EEG Data Using Machine Learning Approach

Xiao-Wei Wang,Dan Nie,Bao-Liang Lu

DOI: https://doi.org/10.1016/j.neucom.2013.06.046

IF: 6

2013-01-01

Neurocomputing

Abstract:Recently, emotion classification from EEG data has attracted much attention with the rapid development of dry electrode techniques, machine learning algorithms, and various real-world applications of brain–computer interface for normal people. Until now, however, researchers had little understanding of the details of relationship between different emotional states and various EEG features. To improve the accuracy of EEG-based emotion classification and visualize the changes of emotional states with time, this paper systematically compares three kinds of existing EEG features for emotion classification, introduces an efficient feature smoothing method for removing the noise unrelated to emotion task, and proposes a simple approach to tracking the trajectory of emotion changes with manifold learning. To examine the effectiveness of these methods introduced in this paper, we design a movie induction experiment that spontaneously leads subjects to real emotional states and collect an EEG data set of six subjects. From experimental results on our EEG data set, we found that (a) power spectrum feature is superior to other two kinds of features; (b) a linear dynamic system based feature smoothing method can significantly improve emotion classification accuracy; and (c) the trajectory of emotion changes can be visualized by reducing subject-independent features with manifold learning.
EEG-Based Identification of Latent Emotional Disorder Using the Machine Learning Approach

Yaling Deng,Fan Wu,Lei Du,Renlai Zhou,Lihong Cao

DOI: https://doi.org/10.1109/ITNEC.2019.8729424

2019-01-01

Abstract:Emotion influences our daily life to a large extent, especially for those who are undergoing bad mood and have high risk for emotional disorders. It is hard to recognize them, but very important so that we can provide intervention before them getting worse. This study used EEG signals to recognize who has high risk for emotional disorders instead of emotion type only. The proposed machine learning method combined the features of multiple cortex areas and frequency bands to find the high risky group for emotional disorders through a kernel SVM classifier. It achieved the accuracy of 95.20%, with all cortex areas and all frequency bands. Results showed that the frontal cortex, central cortex and temporal cortex have a primary influence on identifying emotional disorder and can be used for the reference information for professional diagnose.
EEG-based emotion recognition using frequency domain features and support vector machines

Xiao-Wei Wang,Dan Nie,Bao-Liang Lu

DOI: https://doi.org/10.1007/978-3-642-24955-6_87

2011-01-01

Abstract:Information about the emotional state of users has become more and more important in human-machine interaction and brain-computer interface. This paper introduces an emotion recognition system based on electroencephalogram (EEG) signals. Experiments using movie elicitation are designed for acquiring subject's EEG signals to classify four emotion states, joy, relax, sad, and fear. After pre-processing the EEG signals, we investigate various kinds of EEG features to build an emotion recognition system. To evaluate classification performance, k-nearest neighbor (kNN) algorithm, multilayer perceptron and support vector machines are used as classifiers. Further, a minimum redundancy-maximum relevance method is used for extracting common critical features across subjects. Experimental results indicate that an average test accuracy of 66.51% for classifying four emotion states can be obtained by using frequency domain features and support vector machines.
Electroencephalogram Emotion Recognition Based on Empirical Mode Decomposition and Optimal Feature Selection.

Zhen-Tao Liu,Qiao Xie,Min Wu,Wei-Hua Cao,Dan-Yun Li,Si-Han Li

DOI: https://doi.org/10.1109/tcds.2018.2868121

IF: 4.546

2019-01-01

IEEE Transactions on Cognitive and Developmental Systems

Abstract:Electroencephalogram (EEG) emotion recognition based on a hybrid feature extraction method in empirical mode decomposition domain combining with optimal feature selection based on sequence backward selection is proposed, which can reflect subtle information of multiscale components of unstable and nonlinear EEG signals and remove the reductant features to improve the performance of emotion recognition. The proposal is tested on DEAP dataset, in which the emotional states in the Valance dimension and Arousal dimension are classified by both K-nearest neighbor and support vector machine, respectively. In the experiments, temporal windows of different length and three kinds of rhythms of EEG signal are taken into account for comparison, from which the results show that EEG signal with 1s temporal window achieves highest recognition accuracy of 86.46% in Valence dimension and 84.90% in Arousal dimension, respectively, which is superior to some state-of-the-art works. The proposed method would be applied to real-time emotion recognition in multimodal emotional communication-based humans-robots interaction system.
Emotion Classification from Short-term EEG Signals in Deep Learning

Yanjing Mao,Liang Song,Xinhua Zeng

DOI: https://doi.org/10.1109/ishc54333.2021.00038

2021-01-01

Abstract:Emotion is the result of brain processing after the combination of the surrounding environmental state and the individual physiological state. Its essence is the individual perceptual response to the things around him, and the result of the interaction between man and the environment. There is a strong correlation between human cognitive behavior and psychological activities and EEG signals. The physiological activities of the brain can directly reflect individual emotional activities. EEG signal is one of the important input signals in the affective calculation. Correct emotion representation is a key step in emotion recognition research. This paper aims to study emotion EEG recognition through the representation of emotion classification. The author collected 4000 EEG samples (including 64 channels) of four emotions with a time length of 3s, which were neutral, happy, sad, and fear. According to the characteristic that emotions can not be mutated, setting the sample length of 3s can more flexibly grasp the subjects' emotions in a short time. In this paper, two deep learning methods, LSTM and CNN, are used to identify the positive and negative emotions of the new dataset and DEAP (Database for Emotion Analysis using Physical Signal, including 40 channels). The accuracy of the new dataset is 89% and 74% respectively, which are higher than 72% and 70% of DEAP.
Emotional Feature Extraction and Classification Based on EEG Signals

Changyuan LIU,Wenqiang LI,Xiaojun BI

DOI: https://doi.org/10.3969/j.issn.1004-1699.2019.01.015

2019-01-01

Abstract:As an advanced function of human brain, emotion has a great impact on people's personality characteristics and mental health. By using the online Deap database, emotions are divided according to psychological valence and arousal level, and the two emotions of stress and calm are studied and analyzed. On the basis of using db4 wavelet decomposition and reconstruction algorithm to decompose the signal, according to the characteristics of the asymmetry of EEG signals in the generation of emotions, a new method of emotional feature extraction is proposed, By dividing the differential entropy of right leads by the difference between left and right symmetrical electrodes, and dividing the differential entropy of right leads by the sum of the differential entropy of left and right symmetrical electrodes, the asymmetric entropy characteristics of EEG signals is extracted. Using the support vector machine optimized by genetic algorithm for emotion classification recognition, the average recognition rate is 88.625％.Comparing with the classification recognition rate of traditional features, the classification recognition rate using the asymmetric entropy feature is significantly improved.
A Survey on EEG Based Emotion Recognition

NIE Dan,WANG Xiao-Wei,DUAN Ruo-Nan,LV Bao-Liang

DOI: https://doi.org/10.3969/j.issn.0258-8021.2012.04.018

2012-01-01

Abstract:Emotion is a state that comprehensively represents human feeling,thought and behavior,thus takes an important role in inter-personal human communication.Emotion recognition aims to automatically discriminate different emotional states by using physiological and non-physiological signals acquired from people in order to develop friendly and natural human-machine interface.EEG-based emotion recognition is one of the most common used and efficient methods for emotion recognition.This paper introduced the state-of-the-art techniques,the read-world applications and some critical problems of EEG-based emotion recognition,from the aspects of EEG signals recording,signal preprocessing,feature extraction,feature selection and classification methods.
An Emotional Face Evoked EEG Signal Recognition Method Based on Optimal EEG Feature and Electrodes Selection

Lijuan Duan,Xuebin Wang,Zhen Yang,Haiyan Zhou,Chunpeng Wu,Qi Zhang,Jun Miao

DOI: https://doi.org/10.1007/978-3-642-24955-6_36

2011-01-01

Abstract:In this work, we proposed an emotional face evoked EEG signal recognition framework, within this framework the optimal statistic features were extracted from original signals according to time and space, i.e., the span and electrodes. First, the EEG signals were collected using noise suppression methods, and principal component analysis (PCA) was used to reduce dimension and information redundant of data. Then the optimal statistic features were selected and combined from different electrodes based on the classification performance. We also discussed the contribution of each time span of EEG signals in the same electrodes. Finally, experiments using Fisher, Bayes and SVM classifiers show that our methods offer the better chance for reliable classification of the EEG signal. Moreover, the conclusion is supported by physiological evidence as follows: a) the selected electrodes mainly concentrate in temporal cortex of the right hemisphere, which relates with visual according to previous psychological research; b) the selected time span shows that consciousness of the face picture has a trend from posterior brain regions to anterior brain regions.
Research on the Classification and Channel Selection of Emotional EEG.

Jie Dang,Sirui Wang,Cancheng Li,Yongzong Wang,Hong Peng

DOI: https://doi.org/10.1007/978-3-030-15127-0_2

2019-01-01

Abstract:The emotion affect every aspect of our daily lives. Because of the high temporal resolution and low cost, EEG is widely used in the fields of emotion recognition. This paper studies the effects of different EEG feature combinations and channel selections on emotion recognition. Three effective features including differential entropy (DE), 1st difference (1st) and fractal dimension (FD) were extracted from the EEG signals, and their performances in the three situations of four emotion classifications, two emotions classification on valence and two emotions classification on arousal were calculated and compared by SVM. Two channel selection methods, including the mean relief channel selection algorithm and the common channel selection algorithm, were used to select the best channel. The results showed that when selecting the top 10 channels, the accuracy of the four emotional states classification rate was approximately 95%. This is significant for reducing the number of electrodes and reducing the complexity of brain-computer interface applications in the future.
Machine Learning For Classification Of Antithetical Emotional States

Jeevanshi Sharma,Rajat Maheshwari,Yusuf Uzzaman Khan

DOI: https://doi.org/10.48550/arXiv.2209.02249

IF: 5.414

2022-09-06

Machine Learning

Abstract:Emotion Classification through EEG signals has achieved many advancements. However, the problems like lack of data and learning the important features and patterns have always been areas with scope for improvement both computationally and in prediction accuracy. This works analyses the baseline machine learning classifiers' performance on DEAP Dataset along with a tabular learning approach that provided state-of-the-art comparable results leveraging the performance boost due to its deep learning architecture without deploying heavy neural networks.
Review of the Emotional Feature Extraction and Classification Using EEG Signals

Jiang Wang,Mei Wang

DOI: https://doi.org/10.1016/j.cogr.2021.04.001

2021-01-01

Cognitive Robotics

Abstract:As a subjectively psychological and physiological response to external stimuli, emotion is ubiquitous in our daily life. With the continuous development of the artificial intelligence and brain science, emotion recognition rapidly becomes a multiple discipline research field through EEG signals. This paper investigates the relevantly scientific literature in the past five years and reviews the emotional feature extraction methods and the classification methods using EEG signals. Commonly used feature extraction analysis methods include time domain analysis, frequency domain analysis, and time-frequency domain analysis. The widely used classification methods include machine learning algorithms based on Support Vector Machine (SVM), k-Nearest Neighbor (KNN), Naive Bayes (NB), etc., and their classification accuracy ranges from 57.50% to 95.70%. The classification accuracy of the deep learning algorithms based on Neural Network (NN), Long and Short-Term Memory (LSTM), and Deep Belief Network (DBN) ranges from 63.38% to 97.56%.
Mixed Emotion Recognition Based on EEG Signals

Guanxiong Pei,Bingjie Li,Taihao Li,Cunhang Fan,Chao Zhang,Zhao Lv

DOI: https://doi.org/10.1109/apsipaasc58517.2023.10317209

2023-01-01

Abstract:Advanced emotion monitoring and intervention systems are critical for human-machine cooperation to finish specific tasks in which recognizing mixed emotions is essential. This study is the first to propose intelligent recognition methods of mixed emotions based on EEG signals, thus endowing machines with higher emotional intelligence. We extracted differential entropy (DE), differential asymmetry (DASM), and rational asymmetry (RASM) as features. To better serve the system application and promotion, this study selected three lightweight classifiers, including Random Forest (RF), Probabilistic Neural Networks (PNN), and Generalized Regression Neural Networks (GRNN), all of which achieved good classification performance. The highest classification accuracy is 96.7%. The results of this study can be applied in the fields of mental state assessment, psychological assistance, and clinical psychotherapy.
Machine Learning Approaches for MDD Detection and Emotion Decoding Using EEG Signals

Lijuan Duan,Huifeng Duan,Yuanhua Qiao,Sha,Shunai Qi,Xiaolong Zhang,Juan Huang,Xiaohan Huang,Changming Wang

DOI: https://doi.org/10.3389/fnhum.2020.00284

IF: 3.473

2020-01-01

Frontiers in Human Neuroscience

Abstract:Emotional decoding and automatic identification of major depressive disorder (MDD) are helpful for the timely diagnosis of the disease. Electroencephalography (EEG) is sensitive to changes in the functional state of the human brain, showing its potential to help doctors diagnose MDD. In this paper, an approach for identifying MDD by fusing interhemispheric asymmetry and cross-correlation with EEG signals is proposed and tested on 32 subjects [16 patients with MDD and 16 healthy controls (HCs)]. First, the structural features and connectivity features of the θ-, α-, and β-frequency bands are extracted on the preprocessed and segmented EEG signals. Second, the structural feature matrix of the θ-, α-, and β-frequency bands are added to and subtracted from the connectivity feature matrix to obtain mixed features. Finally, the structural features, connectivity features, and the mixed features are fed to three classifiers to select suitable features for the classification, and it is found that our mode achieves the best classification results using the mixed features. The results are also compared with those from some state-of-the-art methods, and we achieved an accuracy of 94.13%, a sensitivity of 95.74%, a specificity of 93.52%, and an F1-score (f1) of 95.62% on the data from Beijing Anding Hospital, Capital Medical University. The study could be generalized to develop a system that may be helpful in clinical purposes.
A Fine-Grained Approach for EEG-Based Emotion Recognition Using Clustering and Hybrid Deep Neural Networks

Liumei Zhang,Bowen Xia,Yichuan Wang,Wei Zhang,Yu Han

DOI: https://doi.org/10.3390/electronics12234717

IF: 2.9

2023-11-22

Electronics

Abstract:Emotion recognition, as an important part of human-computer interaction, is of great research significance and has already played a role in the fields of artificial intelligence, healthcare, and distance education. In recent times, there has been a growing trend in using deep learning techniques for EEG emotion recognition. These methods have shown higher accuracy in recognizing emotions when compared with traditional machine learning methods. However, most of the current EEG emotion recognition performs multi-category single-label prediction, and is a binary classification problem based on the dimensional model. This simplifies the fact that human emotions are mixed and complex. In order to adapt to real-world applications, fine-grained emotion recognition is necessary. We propose a new method for building emotion classification labels using linguistic resource and density-based spatial clustering of applications with noise (DBSCAN). Additionally, we integrate the frequency domain and spatial features of emotional EEG signals and feed these features into a serial network that combines a convolutional neural network (CNN) and a long short-term memory (LSTM) recurrent neural network (RNN) for EEG emotion feature learning and classification. We conduct emotion classification experiments on the DEAP dataset, and the results show that our method has an average emotion classification accuracy of 92.98% per subject, validating the effectiveness of the improvements we have made to our emotion classification method. Our method for emotion classification holds potential for future use in the domain of affective computing, such as mental health care, education, social media, and so on. By constructing an automatic emotion analysis system using our method to enable the machine to understand the emotional implications conveyed by the subjects' EEG signals, it can provide healthcare professionals with valuable information for effective treatment outcomes.

engineering, electrical & electronic,computer science, information systems,physics, applied
Electroencephalogram Emotion Recognition Using Combined Features in Variational Mode Decomposition Domain

Zhen-Tao Liu,Si-Jun Hu,Jinhua She,Zhaohui Yang,Xin Xu

DOI: https://doi.org/10.1109/tcds.2022.3233858

IF: 4.546

2023-01-01

IEEE Transactions on Cognitive and Developmental Systems

Abstract:Using Electroencephalogram (EEG) to recognize human emotion has attracted increasing attention. However, features extraction from EEG is a challenging work because it is a non-stationary continuous sequential signal. To obtain more pattern information, a combined features extraction method in Variational Mode Decomposition (VMD) domain is proposed, which can extract local features of EEG signals to overcome the effects caused by non-stationarity. This method first decomposes EEG into several components using VMD, and then combined features of Differential Entropy (DE) and Short-Time Energy (STE) are extracted from each component. To optimize combined features, the important features are selected by tree modes, and the feature set is dimensionally reduced by further using Linear Discriminant Analysis (LDA). Moreover, an XGBoost classifier with Bayesian optimization is presented to classify different emotional states. Binary-class and multi-class EEG emotion recognition are conducted on the DEAP dataset, from which the experimental results show that accuracy of binary-class classification is 81.77% for High/Low Valence and 80.47% for High/Low Arousal, and accuracy of 91.41%, 94.27%, 94.27% and 93.49% are obtained for HVHA, LVHA, LVLA and HVLA, respectively, which demonstrate its effectiveness.

robotics,computer science, artificial intelligence,neurosciences
Selection of Optimal EEG Electrodes for Human Emotion Recognition

Jianhua Zhang,Peng Chen

DOI: https://doi.org/10.1016/j.ifacol.2020.12.2753

2020-01-01

IFAC-PapersOnLine

Abstract:In recent years, emotion recognition has attracted increasing interest from researchers from diverse fields. Because of their intrinsic correlation with emotions, physiological signals based emotion recognition method is not susceptible to the so-called social masking and thus more objective than traditional visual, audio or text data based methods. In particular, EEG signals are more responsive to emotion fluctuations than other peripheral physiological signals. In this paper, a 4-class EEG-based emotion classification problem is considered. Firstly the subjective data clustering is performed to identify the optimal number of emotional states. Then wavelet and nonlinear dynamics analyses are used to extract EEG features of emotions. Finally, we consider the brain areas for emotion generation and show that the use of only a small number of EEG electrodes placed on the frontal area of scalp can achieve a 4-class emotion classification accuracy of higher than 90%.
Emotion Analysis on EEG Signal Using Machine Learning and Neural Network

S. M. Masrur Ahmed,Eshaan Tanzim Sabur

2023-07-09

Abstract:Emotion has a significant influence on how one thinks and interacts with others. It serves as a link between how a person feels and the actions one takes, or it could be said that it influences one's life decisions on occasion. Since the patterns of emotions and their reflections vary from person to person, their inquiry must be based on approaches that are effective over a wide range of population regions. To extract features and enhance accuracy, emotion recognition using brain waves or EEG signals requires the implementation of efficient signal processing techniques. Various approaches to human-machine interaction technologies have been ongoing for a long time, and in recent years, researchers have had great success in automatically understanding emotion using brain signals. In our research, several emotional states were classified and tested on EEG signals collected from a well-known publicly available dataset, the DEAP Dataset, using SVM (Support Vector Machine), KNN (K-Nearest Neighbor), and an advanced neural network model, RNN (Recurrent Neural Network), trained with LSTM (Long Short Term Memory). The main purpose of this study is to improve ways to improve emotion recognition performance using brain signals. Emotions, on the other hand, can change with time. As a result, the changes in emotion over time are also examined in our research.

Signal Processing,Artificial Intelligence,Human-Computer Interaction
EEG Signal Classification by Global Field Power

Lijuan Duan,Xuebin Wang,Zhen Yang,Haiyan Zhou,Chunpeng Wu,Qi Zhang,Jun Miao

DOI: https://doi.org/10.4028/www.scientific.net/amm.128-129.1434

2011-01-01

Applied Mechanics and Materials

Abstract:Our project focuses on the emotional face evoked EEG signal recognition. Since EEG signals contain enough information to separate different emotional facial expressions. Thus we propose a new approach which is based on global field power on EEG signal classification. In order to perform this result, firstly, we gather a dataset with EEG signals. This is done by measuring EEG signals from people aged 20-30 that are stimulated by emotional facial expressions (Happy, Neutral, Sad). Secondly, the collected EEG signals are preprocessed through using noise reduction method. And then select features by principal component analysis (PCA) to filter out redundant information. Finally, using fisher classifier and a 10-fold cross validation method for training and testing, a good classification rate is achieved when combination local max global field power EEG signals. The rate is 90.49%.
Dynamic-Attention-based EEG State Transition Modeling for Emotion Recognition

Xinke Shen,Runmin Gan,Kaixuan Wang,Shuyi Yang,Qingzhu Zhang,Quanying Liu,Dan Zhang,Sen Song

2024-11-07

Abstract:Electroencephalogram (EEG)-based emotion decoding can objectively quantify people's emotional state and has broad application prospects in human-computer interaction and early detection of emotional disorders. Recently emerging deep learning architectures have significantly improved the performance of EEG emotion decoding. However, existing methods still fall short of fully capturing the complex spatiotemporal dynamics of neural signals, which are crucial for representing emotion processing. This study proposes a Dynamic-Attention-based EEG State Transition (DAEST) modeling method to characterize EEG spatiotemporal dynamics. The model extracts spatiotemporal components of EEG that represent multiple parallel neural processes and estimates dynamic attention weights on these components to capture transitions in brain states. The model is optimized within a contrastive learning framework for cross-subject emotion recognition. The proposed method achieved state-of-the-art performance on three publicly available datasets: FACED, SEED, and SEED-V. It achieved 75.4% accuracy in the binary classification of positive and negative emotions and 59.3% in nine-class discrete emotion classification on the FACED dataset, 88.1% in the three-class classification of positive, negative, and neutral emotions on the SEED dataset, and 73.6% in five-class discrete emotion classification on the SEED-V dataset. The learned EEG spatiotemporal patterns and dynamic transition properties offer valuable insights into neural dynamics underlying emotion processing.

Human-Computer Interaction,Signal Processing,Neurons and Cognition
EEG-based emotion classification using deep belief networks

Wei-Long Zheng,Jia-Yi Zhu,Yong Peng,Bao-Liang Lu

DOI: https://doi.org/10.1109/ICME.2014.6890166

2014-01-01

Abstract:In recent years, there are many great successes in using deep architectures for unsupervised feature learning from data, especially for images and speech. In this paper, we introduce recent advanced deep learning models to classify two emotional categories (positive and negative) from EEG data. We train a deep belief network (DBN) with differential entropy features extracted from multichannel EEG as input. A hidden markov model (HMM) is integrated to accurately capture a more reliable emotional stage switching. We also compare the performance of the deep models to KNN, SVM and Graph regularized Extreme Learning Machine (GELM). The average accuracies of DBN-HMM, DBN, GELM, SVM, and KNN in our experiments are 87.62%, 86.91%, 85.67%, 84.08%, and 69.66%, respectively. Our experimental results show that the DBN and DBN-HMM models improve the accuracy of EEG-based emotion classification in comparison with the state-of-the-art methods.

Emotional state classification from EEG data using machine learning approach

Emotional State Classification from EEG Data Using Machine Learning Approach

EEG-Based Identification of Latent Emotional Disorder Using the Machine Learning Approach

EEG-based emotion recognition using frequency domain features and support vector machines

Electroencephalogram Emotion Recognition Based on Empirical Mode Decomposition and Optimal Feature Selection.

Emotion Classification from Short-term EEG Signals in Deep Learning

Emotional Feature Extraction and Classification Based on EEG Signals

A Survey on EEG Based Emotion Recognition

An Emotional Face Evoked EEG Signal Recognition Method Based on Optimal EEG Feature and Electrodes Selection

Research on the Classification and Channel Selection of Emotional EEG.

Machine Learning For Classification Of Antithetical Emotional States

Review of the Emotional Feature Extraction and Classification Using EEG Signals

Mixed Emotion Recognition Based on EEG Signals

Machine Learning Approaches for MDD Detection and Emotion Decoding Using EEG Signals

A Fine-Grained Approach for EEG-Based Emotion Recognition Using Clustering and Hybrid Deep Neural Networks

Electroencephalogram Emotion Recognition Using Combined Features in Variational Mode Decomposition Domain

Selection of Optimal EEG Electrodes for Human Emotion Recognition

Emotion Analysis on EEG Signal Using Machine Learning and Neural Network

EEG Signal Classification by Global Field Power

Dynamic-Attention-based EEG State Transition Modeling for Emotion Recognition

EEG-based emotion classification using deep belief networks