NeuroSense: A Novel EEG Dataset Utilizing Low-Cost, Sparse Electrode Devices for Emotion Exploration

Tommaso Colafiglio; Angela Lombardi; Paolo Sorino; Elvira Brattico; Domenico Lofu; Danilo Danese; Eugenio Di Sciascio; Tommaso Di Noia; Fedelucio Narducci

doi:10.1109/ACCESS.2024.3487932

NeuroSense: A Novel EEG Dataset Utilizing Low-Cost, Sparse Electrode Devices for Emotion Exploration

Tommaso Colafiglio, Angela Lombardi^*, Paolo Sorino, Elvira Brattico, Domenico Lofu, Danilo Danese, Eugenio Di Sciascio, Tommaso Di Noia, Fedelucio Narducci

^*Corresponding author for this work

Department of Clinical Medicine - Center for Music In the Brain

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

Abstract

Emotion recognition is crucial in affective computing, aiming to bridge the gap between human emotional states and computer understanding. This study presents NeuroSense, a novel electroencephalography (EEG) dataset utilizing low-cost, sparse electrode devices for emotion exploration. Our dataset comprises EEG signals collected with the portable 4-electrodes device Muse 2 from 30 participants who, thanks to a neurofeedback setting, watch 40 music videos and assess their emotional responses. These assessments use standardized scales gauging arousal, valence, and dominance. Additionally, participants rate their liking for and familiarity with the videos. We develop a comprehensive preprocessing pipeline and employ machine learning algorithms to translate EEG data into meaningful insights about emotional states. We verify the performance of machine learning (ML) models using the NeuroSense dataset. Despite utilizing just 4 electrodes, our models achieve an average accuracy ranging from 75% to 80% across the four quadrants of the dimensional model of emotions. We perform statistical analyses to assess the reliability of the self-reported labels and the classification performance for each participant, identifying potential discrepancies and their implications. We also compare our results with those obtained using other public EEG datasets, highlighting the advantages and limitations of sparse electrode setups in emotion recognition. Our results demonstrate the potential of low-cost EEG devices in emotion recognition, highlighting the effectiveness of ML models in capturing the dynamic nature of emotions. The NeuroSense dataset is publicly available, inviting further research and application in human-computer interaction, mental health monitoring, and beyond.

Original language	English
Journal	IEEE Access
Volume	12
Pages (from-to)	159296-159315
Number of pages	20
ISSN	2169-3536
DOIs	https://doi.org/10.1109/ACCESS.2024.3487932
Publication status	Published - 2024

Keywords

EEG dataset
Emotion recognition
human-computer interaction
low-cost EEG devices
machine learning
Russell's model

Access to Document

10.1109/ACCESS.2024.3487932

Cite this

@article{a320e8e5c9a743e59a4e8dc76c77b732,

title = "NeuroSense: A Novel EEG Dataset Utilizing Low-Cost, Sparse Electrode Devices for Emotion Exploration",

abstract = "Emotion recognition is crucial in affective computing, aiming to bridge the gap between human emotional states and computer understanding. This study presents NeuroSense, a novel electroencephalography (EEG) dataset utilizing low-cost, sparse electrode devices for emotion exploration. Our dataset comprises EEG signals collected with the portable 4-electrodes device Muse 2 from 30 participants who, thanks to a neurofeedback setting, watch 40 music videos and assess their emotional responses. These assessments use standardized scales gauging arousal, valence, and dominance. Additionally, participants rate their liking for and familiarity with the videos. We develop a comprehensive preprocessing pipeline and employ machine learning algorithms to translate EEG data into meaningful insights about emotional states. We verify the performance of machine learning (ML) models using the NeuroSense dataset. Despite utilizing just 4 electrodes, our models achieve an average accuracy ranging from 75% to 80% across the four quadrants of the dimensional model of emotions. We perform statistical analyses to assess the reliability of the self-reported labels and the classification performance for each participant, identifying potential discrepancies and their implications. We also compare our results with those obtained using other public EEG datasets, highlighting the advantages and limitations of sparse electrode setups in emotion recognition. Our results demonstrate the potential of low-cost EEG devices in emotion recognition, highlighting the effectiveness of ML models in capturing the dynamic nature of emotions. The NeuroSense dataset is publicly available, inviting further research and application in human-computer interaction, mental health monitoring, and beyond.",

keywords = "EEG dataset, Emotion recognition, human-computer interaction, low-cost EEG devices, machine learning, Russell's model",

author = "Tommaso Colafiglio and Angela Lombardi and Paolo Sorino and Elvira Brattico and Domenico Lofu and Danilo Danese and {Di Sciascio}, Eugenio and {Di Noia}, Tommaso and Fedelucio Narducci",

note = "Publisher Copyright: {\textcopyright} 2013 IEEE.",

year = "2024",

doi = "10.1109/ACCESS.2024.3487932",

language = "English",

volume = "12",

pages = "159296--159315",

journal = "IEEE Access",

issn = "2169-3536",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - NeuroSense

T2 - A Novel EEG Dataset Utilizing Low-Cost, Sparse Electrode Devices for Emotion Exploration

AU - Colafiglio, Tommaso

AU - Lombardi, Angela

AU - Sorino, Paolo

AU - Brattico, Elvira

AU - Lofu, Domenico

AU - Danese, Danilo

AU - Di Sciascio, Eugenio

AU - Di Noia, Tommaso

AU - Narducci, Fedelucio

PY - 2024

Y1 - 2024

N2 - Emotion recognition is crucial in affective computing, aiming to bridge the gap between human emotional states and computer understanding. This study presents NeuroSense, a novel electroencephalography (EEG) dataset utilizing low-cost, sparse electrode devices for emotion exploration. Our dataset comprises EEG signals collected with the portable 4-electrodes device Muse 2 from 30 participants who, thanks to a neurofeedback setting, watch 40 music videos and assess their emotional responses. These assessments use standardized scales gauging arousal, valence, and dominance. Additionally, participants rate their liking for and familiarity with the videos. We develop a comprehensive preprocessing pipeline and employ machine learning algorithms to translate EEG data into meaningful insights about emotional states. We verify the performance of machine learning (ML) models using the NeuroSense dataset. Despite utilizing just 4 electrodes, our models achieve an average accuracy ranging from 75% to 80% across the four quadrants of the dimensional model of emotions. We perform statistical analyses to assess the reliability of the self-reported labels and the classification performance for each participant, identifying potential discrepancies and their implications. We also compare our results with those obtained using other public EEG datasets, highlighting the advantages and limitations of sparse electrode setups in emotion recognition. Our results demonstrate the potential of low-cost EEG devices in emotion recognition, highlighting the effectiveness of ML models in capturing the dynamic nature of emotions. The NeuroSense dataset is publicly available, inviting further research and application in human-computer interaction, mental health monitoring, and beyond.

AB - Emotion recognition is crucial in affective computing, aiming to bridge the gap between human emotional states and computer understanding. This study presents NeuroSense, a novel electroencephalography (EEG) dataset utilizing low-cost, sparse electrode devices for emotion exploration. Our dataset comprises EEG signals collected with the portable 4-electrodes device Muse 2 from 30 participants who, thanks to a neurofeedback setting, watch 40 music videos and assess their emotional responses. These assessments use standardized scales gauging arousal, valence, and dominance. Additionally, participants rate their liking for and familiarity with the videos. We develop a comprehensive preprocessing pipeline and employ machine learning algorithms to translate EEG data into meaningful insights about emotional states. We verify the performance of machine learning (ML) models using the NeuroSense dataset. Despite utilizing just 4 electrodes, our models achieve an average accuracy ranging from 75% to 80% across the four quadrants of the dimensional model of emotions. We perform statistical analyses to assess the reliability of the self-reported labels and the classification performance for each participant, identifying potential discrepancies and their implications. We also compare our results with those obtained using other public EEG datasets, highlighting the advantages and limitations of sparse electrode setups in emotion recognition. Our results demonstrate the potential of low-cost EEG devices in emotion recognition, highlighting the effectiveness of ML models in capturing the dynamic nature of emotions. The NeuroSense dataset is publicly available, inviting further research and application in human-computer interaction, mental health monitoring, and beyond.

KW - EEG dataset

KW - Emotion recognition

KW - human-computer interaction

KW - low-cost EEG devices

KW - machine learning

KW - Russell's model

UR - http://www.scopus.com/inward/record.url?scp=85208216724&partnerID=8YFLogxK

U2 - 10.1109/ACCESS.2024.3487932

DO - 10.1109/ACCESS.2024.3487932

M3 - Journal article

AN - SCOPUS:85208216724

SN - 2169-3536

VL - 12

SP - 159296

EP - 159315

JO - IEEE Access

JF - IEEE Access

ER -

NeuroSense: A Novel EEG Dataset Utilizing Low-Cost, Sparse Electrode Devices for Emotion Exploration

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this