To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
A Review on Data Scarcity in EEG-based Emotion Recognition |
|
Author |
Maryam Al Dabel |
| Citation |
Vol. 25 No. 4 pp. 238-248 |
|
Abstract |
Emotion recognition through electroencephalography (EEG) has become a vital area of research, offering significant applications in psychology, human-computer interaction, and affective computing. However, the effectiveness of EEG-based emotion recognition systems is often compromised by data scarcity, characterized by limited sample sizes and variability in emotional expressions across individuals. This review examines the challenges posed by data scarcity in EEG studies, highlighting its impact on model performance, generalizability, and research credibility. We explore various techniques aimed at addressing these challenges, including data augmentation, synthetic data generation through Generative Adversarial Networks (GANs), transfer learning, cross-dataset validation, and collaborative data sharing. Recent advancements in deep learning, novel signal processing methods, and the integration of multimodal approaches and artificial intelligence are also discussed, showcasing their potential to enhance emotion recognition capabilities. The review emphasizes the need for larger, more diverse datasets and interdisciplinary collaboration to advance the field. By addressing data scarcity and embracing innovative methodologies, the future of EEG-based emotion recognition holds promising avenues for improved mental health assessments and enhanced user experiences across various applications. |
|
Keywords |
EEG (Electroencephalography), Emotion Recognition, Data Scarcity, Machine Learning. |
|
URL |
