scholarly journals Four-classes Human Emotion Recognition Via Entropy Characteristic and Random Forest。

2020 ◽  
Vol 49 (3) ◽  
pp. 285-298
Author(s):  
Jian Zhang ◽  
Yihou Min
Keyword(s):  
Random Forest ◽  
Emotion Recognition ◽  
Classification Accuracy ◽  
Wavelet Packet ◽  
Recognition Rate ◽  
Eeg Signal ◽  
Differential Entropy ◽  
Eeg Signals ◽  
Human Emotion ◽  
Before And After

Human Emotion Recognition is of vital importance to realize human-computer interaction (HCI), while multichannel electroencephalogram (EEG) signals gradually replace other physiological signals and become the main basis of emotional recognition research with the development of brain-computer interface (BCI). However, the accuracy of emotional classification based on EEG signals under video stimulation is not stable, which may be related to the characteristics of  EEG signals before receiving stimulation. In this study, we extract the change of Differential Entropy (DE) before and after stimulation based on wavelet packet transform (WPT) to identify individual emotional state. Using the EEG emotion database DEAP, we divide the experimental EEG data in the database equally into 15 sets and extract their differential entropy on the basis of WPT. Then we calculate value of DE change of each separated EEG signal set. Finally, we divide the emotion into four categories in the two-dimensional valence-arousal emotional space by combining it with the integrated algorithm, Random Forest (RF). The simulation results show that the WPT-RF model established by this method greatly improves the recognition rate of EEG signal, with an average classification accuracy of 87.3%. In addition, we use WPT-RF model to train individual subjects, and the classification accuracy reached 97.7%.

CLASSIFICATION OF ICTAL EEG USING MODELING BASED SPECTRAL AND TEMPORAL FEATURES ON INSTANTANEOUS AMPLITUDE-FREQUENCY COMPONENTS OF IMFs

2018 ◽  
Vol 30 (06) ◽  
pp. 1850042 ◽  
Author(s):  
K. S. Biju ◽  
M. G. Jibukumar
Keyword(s):  
Spectral Density ◽  
Power Spectral Density ◽  
Classification Accuracy ◽  
Spectral Features ◽  
Eeg Signal ◽  
Instantaneous Amplitude ◽  
Eeg Signals ◽  
Hilbert Huang Transform ◽  
Temporal Features ◽  
Power Spectral

In the present study, a method for classifying the different ictal stages in electroencephalogram (EEG) signals is proposed. The main symptoms of epilepsy are indicated by ictal activities, which trigger widespread neurological disorders other than stroke and thus affect the world population. In this work, a novel ictal classification method that combines the spectral and temporal features of twin components in Hilbert–Huang transform is proposed. Spectral features of instantaneous amplitude (IA) function are obtained based on the power spectral density of autoregressive (AR) modeling. Here four different cases of ictal activities of EEG signal are classified. In each case first and second intrinsic mode function of Hilbert–Huang transform are tabulated. The power spectral density of AR(6) and AR(10) model are done for IA1 and IA2 components of each case. Temporal features of either instantaneous frequency (IF) function or IA are computed. The feature vectors are tested in a well-known database of different classes in interictal, ictal, and normal activities of EEG signals. The discriminating power of each vector is evaluated through one-way analysis of variance, and the classification results are verified using an artificial neural network (ANN) classifier. The performance of the classifier was assessed in term of sensitivity, specificity, and total classification accuracy. The spectral features of the AR(10) of IA and the temporal features of IA yielded 100% accuracy, 100% sensitivity, and 100% specificity in the ictal classification. By contrast, these features obtained only 83.33% of the total classification accuracy in ictal and interictal EEG signal.

scholarly journals Sub-band selection approach to artifact suppression from electroencephalography signal using hybrid wavelet transform

2021 ◽  
Vol 18 (1) ◽  
pp. 172988142199226
Author(s):  
Jannatul Ferdous ◽  
Sujan Ali ◽  
Ekramul Hamid ◽  
Khademul Islam Molla
Keyword(s):  
Wavelet Transform ◽  
Performance Metrics ◽  
Wavelet Packet ◽  
Reference Signal ◽  
Computational Cost ◽  
Discrete Wavelet ◽  
Eeg Signal ◽  
Eeg Signals ◽  
Finite Set ◽  
Artifact Suppression

This article presents a hybrid wavelet-based algorithm to suppress the ocular artifacts from electroencephalography (EEG) signals. The hybrid wavelet transform (HWT) method is designed by the combination of discrete wavelet decomposition and wavelet packet transform. The artifact suppression is performed by the selection of sub-bands obtained by HWT. Fractional Gaussian noise (fGn) is used as the reference signal to select the sub-bands containing the artifacts. The multichannel EEG signal is decomposed HWT into a finite set of sub-bands. The energies of the sub-bands are compared to that of the fGn to the desired sub-band signals. The EEG signal is reconstructed by the selected sub-bands consisting of EEG. The experiments are conducted for both simulated and real EEG signals to study the performance of the proposed algorithm. The results are compared with recently developed algorithms of artifact suppression. It is found that the proposed method performs better than the methods compared in terms of performance metrics and computational cost.

scholarly journals Deep Learning-based Categorical and Dimensional Emotion Recognition for Written and Spoken Text

2019 ◽  
Author(s):  
Bagus Tris Atmaja
Keyword(s):  
Deep Learning ◽  
Emotion Recognition ◽  
Unsupervised Learning ◽  
Recognition Rate ◽  
Product Reviews ◽  
Text Corpus ◽  
Human Emotion ◽  
Written Text ◽  
Spoken Text ◽  
Speech Transcription

The demand for recognizing emotion in text has grown increasingly as human emotion can be expressed via text and manytechnologies, such as product reviews and speech transcription, can benefit from text emotion recognition. The study of text emotionrecognition was established some decades ago using unsupervised learning and a small amount of data. Advancements incomputation hardware and in the development of larger text corpus have enabled us to analyze emotion in the text by moresophisticated techniques. This paper presents a deep learning-based approach for the recognition of categorical and dimensionalemotion from both written and spoken texts. The result shows that the system performs better on both categorical and dimensional task( > 60% accuracy and < 20% error) with a larger dataset compared to a smaller dataset. We also found the recognition rate is affectedby both the size of the data and the number of emotion categories. On the dimensional task, a larger amount of data consistentlyprovided a better result. Recognition of categorical emotion on a spoken text is easier than on a written text, while on dimensional task,the written text yielded better performance.

scholarly journals GameEmo-CapsNet: Emotion Recognition from Single-Channel EEG Signals Using the 1D Capsule Networks

2021 ◽  
Vol 38 (6) ◽  
pp. 1689-1698
Author(s):  
Suat Toraman ◽  
Ömer Osman Dursun
Keyword(s):  
Emotion Recognition ◽  
Facial Expressions ◽  
Classification Accuracy ◽  
Single Channel ◽  
Visual Stimuli ◽  
Classification Performance ◽  
Eeg Signals ◽  
Machine Learning Methods ◽  
First Time ◽  
Better Than

Human emotion recognition with machine learning methods through electroencephalographic (EEG) signals has become a highly interesting subject for researchers. Although it is simple to define emotions that can be expressed physically such as speech, facial expressions, and gestures, it is more difficult to define psychological emotions that are expressed internally. The most important stimuli in revealing inner emotions are aural and visual stimuli. In this study, EEG signals using both aural and visual stimuli were examined and emotions were evaluated in both binary and multi-class emotion recognitions models. A general emotion recognition model was proposed for non-subject-based classification. Unlike in previous studies, a subject-based testing was performed for the first time in the literature. Capsule Networks, a new neural network model, has been developed for binary and multi-class emotion recognition. In the proposed method, a novel fusion strategy was introduced for binary-class emotion recognition and the model was tested using the GAMEEMO dataset. Binary-class emotion recognition achieved a classification accuracy which was 10% better than the classification performance achieved in other studies in the literature. Based on these findings, we suggest that the proposed method will bring a different perspective to emotion recognition.

scholarly journals Emotion recognition using electroencephalogram signal

2019 ◽  
Vol 15 (2) ◽  
pp. 786
Author(s):  
Aqila Nur Nadira Mohammad Yosi ◽  
Khairul Azami Sidek ◽  
Hamwira Sakti Yaacob ◽  
Marini Othman ◽  
Ahmad Zamani Jusoh
Keyword(s):  
High Resolution ◽  
Emotion Recognition ◽  
Brain Activity ◽  
Eeg Signal ◽  
Eeg Signals ◽  
Recognition Mechanism ◽  
High Resolution Data ◽  
Average Percentage ◽  
Low Pass ◽  
Acquisition Device

<p class="Abstract">Emotion play an essential role in human’s life and it is not consciously controlled. Some of the emotion can be easily expressed by facial expressions, speech, behavior and gesture but some are not. This study investigates the emotion recognition using electroencephalogram (EEG) signal. Undoubtedly, EEG signals can detect human brain activity accurately with high resolution data acquisition device as compared to other biological signals. Changes in the human brain’s electrical activity occur very quickly, thus a high resolution device is required to determine the emotion precisely. In this study, we will prove the strength and reliability of EEG signals as an emotion recognition mechanism for four different emotions which are happy, sad, fear and calm. Data of six different subjects were collected by using BrainMarker EXG device which consist of 19 channels. The pre-processing stage was performed using second order of low pass Butterworth filter to remove the unwanted signals. Then, two ranges of frequency bands were extracted from the signals which are alpha and beta. Finally, these samples will be classified using MLP Neural Network. Classification accuracy up to 91% is achieved and the average percentage of accuracy for calm, fear, happy and sad are 83.5%, 87.3%, 85.83% and 87.6% respectively. Thus, a proof of concept, this study has been capable of proposing a system of recognizing four states of emotion which are happy, sad, fear and calm by using EEG signal.</p>

scholarly journals Human Emotion Recognition Based on EEG Signal Using Fast Fourier Transform and K-Nearest Neighbor

2020 ◽  
Vol 5 (6) ◽  
pp. 1082-1088
Author(s):  
Anton Yudhana ◽  
Akbar Muslim ◽  
Dewi Eko Wati ◽  
Intan Puspitasari ◽  
Ahmad Azhari ◽  
...  
Keyword(s):  
Fourier Transform ◽  
Fast Fourier Transform ◽  
Emotion Recognition ◽  
Nearest Neighbor ◽  
Eeg Signal ◽  
K Nearest Neighbor ◽  
Human Emotion

Research of Classification Methods of EEG Signal Based on Wavelet Packet Transform and LVQ Neural

2014 ◽  
Vol 1049-1050 ◽  
pp. 1626-1630
Author(s):  
Xue Yan Pang ◽  
Shi Nin Yin ◽  
Hong Zhou Li ◽  
Jian Ming Zhu ◽  
Zhen Cheng Chen
Keyword(s):  
Neural Network ◽  
Network Model ◽  
Neural Network Model ◽  
Wavelet Packet ◽  
Recognition Rate ◽  
Signal Recognition ◽  
Eeg Signals ◽  
Specific Frequency ◽  
Correct Recognition Rate ◽  
Electroencephalogram Eeg

In order to improve the correct recognition rate of EEG(Electroencephalogram,EEG) signals to meet the needs of Brain-Computer Interface system,this paper put forward a new method of signal recognition which combines wavelet packet decomposition and LVQ neural network.First,using the method of wavelet packet to analyze the signal,and then extract the specific frequency band’s energy of wavelet packet as characteristics.Then using the LVQ neural network model to study the distinguishing between the two EEG datas of Motor Imagery.The simulation experiment uses Matlab software to design LVQ neural network model to judge the two kinds of Motor Imagery task.In the process of judgment,respecti-vely to classify the data by using BP neural network and LVQ neural network.Experimental results show that the LVQ neural network can have a higher correct accuracy to recognize the motor imaginary task than BP neural.

Sign in / Sign up

Export Citation Format

Share Document