Abstract
A number of previous works have adopted a subject independent approach for recognizing emotions from Electroencephalography (EEG) signals, and attempted to build a global model by treating data from different subjects as if they belong to the same individual. In this paper we visually explore the data provided in four different standard datasets when using Power Spectral Density features, and show that the subject-dependent component in the EEG signal is far stronger than the emotion-related component. In addition, the session-dependency that is also found discourages the application of this type of features from EEG signals in a biometric context.
Original language | English |
---|---|
Title of host publication | 33rd Annual European Simulation and Modelling Conference 2019, ESM 2019 |
Editors | Pilar Fuster-Parra, Oscar Valero Sierra |
Publisher | EUROSIS |
Pages | 229-233 |
Number of pages | 5 |
ISBN (Electronic) | 9789492859099 |
Publication status | Published - 2019 |
Externally published | Yes |
Event | 33rd Annual European Simulation and Modelling Conference, ESM 2019 - Plama de Mallorca, Spain Duration: 28 Oct 2019 → 30 Oct 2019 |
Conference
Conference | 33rd Annual European Simulation and Modelling Conference, ESM 2019 |
---|---|
Country/Territory | Spain |
City | Plama de Mallorca |
Period | 28/10/19 → 30/10/19 |
Bibliographical note
Funding Information:This research has been partly supported by CECOTEC INNOVACIONES, S.L. and the Spanish Ministry of Economy and Competitiveness through projects TIN2014-59641-C2-1-P and PGC2018-096463-B-I00.
Publisher Copyright:
Copyright © 2019 EUROSIS-ETI.
Keywords
- Affect
- Biometrics
- EEG
- Emotion modeling