We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.HC

Change to browse by:

cs

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Human-Computer Interaction

Title: BEAMERS: Brain-Engaged, Active Music-based Emotion Regulation System

Abstract: With the increasing demands of emotion comprehension and regulation in our daily life, a customized music-based emotion regulation system is introduced by employing current EEG information and song features, which predicts users' emotion variation in the valence-arousal model before recommending music. The work shows that: (1) a novel music-based emotion regulation system with a commercial EEG device is designed without employing deterministic emotion recognition models for daily usage; (2) the system considers users' variant emotions towards the same song, and by which calculate user's emotion instability and it is in accordance with Big Five Personality Test; (3) the system supports different emotion regulation styles with users' designation of desired emotion variation, and achieves an accuracy of over $0.85$ with 2-seconds EEG data; (4) people feel easier to report their emotion variation comparing with absolute emotional states, and would accept a more delicate music recommendation system for emotion regulation according to the questionnaire.
Subjects: Human-Computer Interaction (cs.HC)
Cite as: arXiv:2211.14609 [cs.HC]
  (or arXiv:2211.14609v1 [cs.HC] for this version)

Submission history

From: Jiyang Li [view email]
[v1] Sat, 26 Nov 2022 16:37:13 GMT (1751kb,D)

Link back to: arXiv, form interface, contact.