Name
Title | ||
|---|---|---|
Recognition of persisting emotional valence from EEG using convolutional neural networks |
Abstract | ||
|---|---|---|
Recently there has been considerable interest in EEG-based emotion recognition (EEG-ER), which is one of the utilization of BCI. However, it is not easy to realize the EEG-ER system which can recognize emotions with high accuracy because of the tendency for important information in EEG signals to be concealed by noises. Deep learning is the golden tool to grasp the features concealed in EEG data and enable highly accurate EEG-ER because deep neural networks (DNNs) may have higher recognition capability than humans'. The publicly available dataset named DEAP, which is for emotion analysis using EEG, was used in the experiment. The CNN and a conventional model used for comparison are evaluated by the tests according to 11-fold cross validation scheme. EEG raw data obtained from 16 electrodes without general preprocesses were used as input data. The models classify and recognize EEG signals according to the emotional states "positive" or "negative" which were caused by watching music videos. The results show that the more training data are, the much higher the accuracies of CNNs are (by over 20%). It also suggests that the increased training data need not to belong to the same person's EEG data as the test data so as to get the CNN recognizing emotions accurately. The results indicate that there are not only the considerable amount of the interpersonal difference but also commonality of EEG properties. |
Year | DOI | Venue |
|---|---|---|
2016 | 10.1109/IWCIA.2016.7805744 | 2016 IEEE 9th International Workshop on Computational Intelligence and Applications (IWCIA) |
Keywords | Field | DocType |
BCI,EEG,Emotion recognition,Convolutional neural networks,Deep learning,Interpersonal difference/commonality | Convolutional neural network,Computer science,Brain–computer interface,Raw data,Artificial intelligence,Deep learning,Electroencephalography,Computer vision,GRASP,Speech recognition,Test data,Cross-validation,Machine learning | Conference |
ISBN | Citations | PageRank |
978-1-5090-2776-7 | 0 | 0.34 |
References | Authors | |
5 | 2 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Miku Yanagimoto | 1 | 0 | 0.68 |
| Sugimoto, C. | 2 | 8 | 7.06 |