Version 1
: Received: 24 December 2019 / Approved: 25 December 2019 / Online: 25 December 2019 (09:26:52 CET)
Version 2
: Received: 31 December 2019 / Approved: 31 December 2019 / Online: 31 December 2019 (10:05:27 CET)
How to cite:
Bilucaglia, M.; Duma, G.M.; Mento, G.; Semenzato, L.; Tressoldi, P. EEG Signal Classification: An Application to the Emotion Related Brain Anticipatory Activity. Preprints2019, 2019120340
Bilucaglia, M.; Duma, G.M.; Mento, G.; Semenzato, L.; Tressoldi, P. EEG Signal Classification: An Application to the Emotion Related Brain Anticipatory Activity. Preprints 2019, 2019120340
Bilucaglia, M.; Duma, G.M.; Mento, G.; Semenzato, L.; Tressoldi, P. EEG Signal Classification: An Application to the Emotion Related Brain Anticipatory Activity. Preprints2019, 2019120340
APA Style
Bilucaglia, M., Duma, G.M., Mento, G., Semenzato, L., & Tressoldi, P. (2019). EEG Signal Classification: An Application to the Emotion Related Brain Anticipatory Activity. Preprints. https://doi.org/
Chicago/Turabian Style
Bilucaglia, M., Luca Semenzato and Patrizio Tressoldi. 2019 "EEG Signal Classification: An Application to the Emotion Related Brain Anticipatory Activity" Preprints. https://doi.org/
Abstract
Machine Learning (ML) approaches have been fruitfully applied to several classification problems of neurophysiological activity. Considering the relevance of emotion in human cognition and behaviour, ML found an important application field in emotion identification based on neurophysiological activity. Nonetheless, the literature results present a high variability depending on the neuronal activity measurement, the signal features and the classifier type. The present work aims to provide new methodological insight on ML applied to emotion identification based on electrophysiological brain activity. For this reason, we recorded EEG activity while emotional stimuli, high and low arousal (auditory and visual) were provided to a group of healthy participants. Our target signal to classify was the pre-stimulus onset brain activity. Classification performance of three different classifiers (LDA, SVM and kNN) was compared using both spectral and temporal features. Furthermore, we also contrasted the classifiers performance with static and dynamic (time evolving) features. The results show a clear increased in classification accuracy with temporal dynamic features. In particular, the SVM classifiers with temporal features showed the best accuracy (63.8 %) in classifying high vs. low arousal auditory stimuli.
Keywords
emotion recognition; EEG signal decoding; brain anticipatory activity; machine learning; emotion related brain activity
Subject
Social Sciences, Behavior Sciences
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Commenter: Patrizio Tressoldi
Commenter's Conflict of Interests: Author
Added more details about the EEG database in the Stimuli and experimental paradigm paragraph