Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Emotion Recognition Based on EEG Signals Using LIBSVM as the Classifier
In order to improve the emotion recognition rate, this paper proposes an electroencephalograph (EEG) emotion recognition model using a library for support vector machine (LIBSVM) as the classifier. In this paper, we collected EEG signals from 10 volunteers along with the participants’ self-assessment of their affective state after each stimulus, in terms of Valence and arousal. After these signals are filtered, we calculate the features of Lempel-Ziv complexity, wavelet detail coefficient, and the co-integration relationship degree first. At the same time, EMD is carried out to calculate the average approximate entropy of the first four Intrinsic Mode Functions (IMFs). At last, all the features extracted will input into the LIBSVM for training and testing, and complete emotion recognition. In this paper, two classifications are carried out on the two dimensions of Valence and Arousal, respectively. The experimental results show that the average emotional recognition rate is 83.64% and 75.11%, respectively, which proves that the proposed scheme has a certain feasibility.
Emotion Recognition Based on EEG Signals Using LIBSVM as the Classifier
In order to improve the emotion recognition rate, this paper proposes an electroencephalograph (EEG) emotion recognition model using a library for support vector machine (LIBSVM) as the classifier. In this paper, we collected EEG signals from 10 volunteers along with the participants’ self-assessment of their affective state after each stimulus, in terms of Valence and arousal. After these signals are filtered, we calculate the features of Lempel-Ziv complexity, wavelet detail coefficient, and the co-integration relationship degree first. At the same time, EMD is carried out to calculate the average approximate entropy of the first four Intrinsic Mode Functions (IMFs). At last, all the features extracted will input into the LIBSVM for training and testing, and complete emotion recognition. In this paper, two classifications are carried out on the two dimensions of Valence and Arousal, respectively. The experimental results show that the average emotional recognition rate is 83.64% and 75.11%, respectively, which proves that the proposed scheme has a certain feasibility.
Emotion Recognition Based on EEG Signals Using LIBSVM as the Classifier
Stud. in Distributed Intelligence
Yuan, Xiaohui (Herausgeber:in) / Elhoseny, Mohamed (Herausgeber:in) / Chen, Tian (Autor:in) / Ju, Sihang (Autor:in) / Ren, Fuji (Autor:in) / Fan, Mingyan (Autor:in) / An, Xin (Autor:in)
26.06.2020
13 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
| DOAJ | 2021
Emotion Recognition Using Speaker Cues
| IEEE | 2020
Speech Emotion Recognition Based on EMD in Noisy Environments
| British Library Conference Proceedings | 2014