Emotion Recognition Using Text Embedding Models: Wearable and Wireless EEG Without Fixed EEG Channel Configurations

Quoc-Toan Nguyen; Zheng Huiru; Tahia Tazin; Linh Le; Tuan L. Vo; Nhu-Tri Tran; David Williams-King; Benjamin Tag

doi:10.1145/3708319.3733645

Emotion Recognition Using Text Embedding Models: Wearable and Wireless EEG Without Fixed EEG Channel Configurations

Quoc-Toan Nguyen
, Zheng Huiru
, Tahia Tazin
, Linh Le
, Tuan L. Vo
, Nhu-Tri Tran
, David Williams-King
, Benjamin Tag

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Citation (Scopus)

Abstract

Emotion recognition methods using Artificial Intelligence (AI) and wearable/wireless Electroencephalography (wEEG) are promising, as wEEG signals effectively and conveniently capture brain activities related to emotions. However, conventional AI models require separate development for each wEEG channel configuration, limiting adaptability and increasing costs. To address this gap, this paper proposes a framework for leveraging text embedding models to transform wEEG signals into a standardised representation for different wEEG channel setups to be compatible with a single AI model. This approach enhances scalability, adaptability, and resource efficiency, making AI-driven emotion recognition more cost-effective and accessible. Our proposed method achieves an accuracy of 0.9368 and 0.9484 with snowflake-arctic-embed-l-v2.0 with 2-second epoching and multilingual-e5-large-instruct using 5-second epoching. This proposed method can be effectively applied across various wEEG channel configurations to support tasks to improve or explore human well-being, such as stress monitoring or emotion self-regulation.

Original language	English
Title of host publication	UMAP 2025 - Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization
Publisher	Association for Computing Machinery
Pages	476-488
Number of pages	13
ISBN (Electronic)	9798400713996
DOIs	https://doi.org/10.1145/3708319.3733645
Publication status	Published online - 12 Jun 2025
Event	UMAP Adjunct '25: Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization - New York City, United States Duration: 16 Jun 2025 → 19 Jun 2025

Publication series

Name	UMAP 2025 - Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization

Conference

Conference	UMAP Adjunct '25: Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization
Country/Territory	United States
City	New York City
Period	16/06/25 → 19/06/25

Bibliographical note

Publisher Copyright:
© 2025 Copyright held by the owner/author(s). Publication rights licensed to ACM.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 8 Decent Work and Economic Growth

Keywords

Emotion Recognition
EEG
Text Embedding Model
Affective Computing
Brain-Computer Interface
wEEG
aBCI
HCI
Brain-Computer Interface

Access to Document

10.1145/3708319.3733645

Cite this

Nguyen, Q.-T., Huiru, Z., Tazin, T., Le, L., Vo, T. L., Tran, N.-T., Williams-King, D., & Tag, B. (2025). Emotion Recognition Using Text Embedding Models: Wearable and Wireless EEG Without Fixed EEG Channel Configurations. In UMAP 2025 - Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization (pp. 476-488). (UMAP 2025 - Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization). Association for Computing Machinery. Advance online publication. https://doi.org/10.1145/3708319.3733645

Nguyen, Quoc-Toan ; Huiru, Zheng ; Tazin, Tahia et al. / Emotion Recognition Using Text Embedding Models: Wearable and Wireless EEG Without Fixed EEG Channel Configurations. UMAP 2025 - Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization. Association for Computing Machinery, 2025. pp. 476-488 (UMAP 2025 - Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization).

@inproceedings{2e37dd4cf416451891b7efa357dae5e6,

title = "Emotion Recognition Using Text Embedding Models: Wearable and Wireless EEG Without Fixed EEG Channel Configurations",

abstract = "Emotion recognition methods using Artificial Intelligence (AI) and wearable/wireless Electroencephalography (wEEG) are promising, as wEEG signals effectively and conveniently capture brain activities related to emotions. However, conventional AI models require separate development for each wEEG channel configuration, limiting adaptability and increasing costs. To address this gap, this paper proposes a framework for leveraging text embedding models to transform wEEG signals into a standardised representation for different wEEG channel setups to be compatible with a single AI model. This approach enhances scalability, adaptability, and resource efficiency, making AI-driven emotion recognition more cost-effective and accessible. Our proposed method achieves an accuracy of 0.9368 and 0.9484 with snowflake-arctic-embed-l-v2.0 with 2-second epoching and multilingual-e5-large-instruct using 5-second epoching. This proposed method can be effectively applied across various wEEG channel configurations to support tasks to improve or explore human well-being, such as stress monitoring or emotion self-regulation.",

keywords = "Emotion Recognition, EEG, Text Embedding Model, Affective Computing, Brain-Computer Interface, wEEG, aBCI, HCI, Brain-Computer Interface",

author = "Quoc-Toan Nguyen and Zheng Huiru and Tahia Tazin and Linh Le and Vo, \{Tuan L.\} and Nhu-Tri Tran and David Williams-King and Benjamin Tag",

note = "Publisher Copyright: {\textcopyright} 2025 Copyright held by the owner/author(s). Publication rights licensed to ACM.; UMAP Adjunct '25: Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization ; Conference date: 16-06-2025 Through 19-06-2025",

year = "2025",

month = jun,

day = "12",

doi = "10.1145/3708319.3733645",

language = "English",

series = "UMAP 2025 - Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization",

publisher = "Association for Computing Machinery",

pages = "476--488",

booktitle = "UMAP 2025 - Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization",

address = "United States",

}

Nguyen, Q-T, Huiru, Z, Tazin, T, Le, L, Vo, TL, Tran, N-T, Williams-King, D & Tag, B 2025, Emotion Recognition Using Text Embedding Models: Wearable and Wireless EEG Without Fixed EEG Channel Configurations. in UMAP 2025 - Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization. UMAP 2025 - Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization, Association for Computing Machinery, pp. 476-488, UMAP Adjunct '25: Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization, New York City, New York, United States, 16/06/25. https://doi.org/10.1145/3708319.3733645

Emotion Recognition Using Text Embedding Models: Wearable and Wireless EEG Without Fixed EEG Channel Configurations. / Nguyen, Quoc-Toan; Huiru, Zheng; Tazin, Tahia et al.
UMAP 2025 - Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization. Association for Computing Machinery, 2025. p. 476-488 (UMAP 2025 - Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Emotion Recognition Using Text Embedding Models: Wearable and Wireless EEG Without Fixed EEG Channel Configurations

AU - Nguyen, Quoc-Toan

AU - Huiru, Zheng

AU - Tazin, Tahia

AU - Le, Linh

AU - Vo, Tuan L.

AU - Tran, Nhu-Tri

AU - Williams-King, David

AU - Tag, Benjamin

PY - 2025/6/12

Y1 - 2025/6/12

N2 - Emotion recognition methods using Artificial Intelligence (AI) and wearable/wireless Electroencephalography (wEEG) are promising, as wEEG signals effectively and conveniently capture brain activities related to emotions. However, conventional AI models require separate development for each wEEG channel configuration, limiting adaptability and increasing costs. To address this gap, this paper proposes a framework for leveraging text embedding models to transform wEEG signals into a standardised representation for different wEEG channel setups to be compatible with a single AI model. This approach enhances scalability, adaptability, and resource efficiency, making AI-driven emotion recognition more cost-effective and accessible. Our proposed method achieves an accuracy of 0.9368 and 0.9484 with snowflake-arctic-embed-l-v2.0 with 2-second epoching and multilingual-e5-large-instruct using 5-second epoching. This proposed method can be effectively applied across various wEEG channel configurations to support tasks to improve or explore human well-being, such as stress monitoring or emotion self-regulation.

AB - Emotion recognition methods using Artificial Intelligence (AI) and wearable/wireless Electroencephalography (wEEG) are promising, as wEEG signals effectively and conveniently capture brain activities related to emotions. However, conventional AI models require separate development for each wEEG channel configuration, limiting adaptability and increasing costs. To address this gap, this paper proposes a framework for leveraging text embedding models to transform wEEG signals into a standardised representation for different wEEG channel setups to be compatible with a single AI model. This approach enhances scalability, adaptability, and resource efficiency, making AI-driven emotion recognition more cost-effective and accessible. Our proposed method achieves an accuracy of 0.9368 and 0.9484 with snowflake-arctic-embed-l-v2.0 with 2-second epoching and multilingual-e5-large-instruct using 5-second epoching. This proposed method can be effectively applied across various wEEG channel configurations to support tasks to improve or explore human well-being, such as stress monitoring or emotion self-regulation.

KW - Emotion Recognition

KW - EEG

KW - Text Embedding Model

KW - Affective Computing

KW - Brain-Computer Interface

KW - wEEG

KW - aBCI

KW - HCI

KW - Brain-Computer Interface

UR - https://www.scopus.com/pages/publications/105011070651

U2 - 10.1145/3708319.3733645

DO - 10.1145/3708319.3733645

M3 - Conference contribution

T3 - UMAP 2025 - Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization

SP - 476

EP - 488

BT - UMAP 2025 - Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization

PB - Association for Computing Machinery

T2 - UMAP Adjunct '25: Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization

Y2 - 16 June 2025 through 19 June 2025

ER -

Nguyen QT, Huiru Z, Tazin T, Le L, Vo TL, Tran NT et al. Emotion Recognition Using Text Embedding Models: Wearable and Wireless EEG Without Fixed EEG Channel Configurations. In UMAP 2025 - Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization. Association for Computing Machinery. 2025. p. 476-488. (UMAP 2025 - Adjunct Proceedings of the 33rd ACM Conference on User Modeling, Adaptation and Personalization). Epub 2025 Jun 12. doi: 10.1145/3708319.3733645

Emotion Recognition Using Text Embedding Models: Wearable and Wireless EEG Without Fixed EEG Channel Configurations

Abstract

Publication series

Conference

Bibliographical note

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this