Quantum Computational Modeling for Affective Assessment in Virtual Reality Systems

This research explores the use of quantum machine learning-more specifically, the Quantum Support Vector Machine (QSVM)-to predict emotional reactions triggered in virtual reality (VR) settings. The study's experimental setup involved provoking different emotional states via VR gameplay element...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2024 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW) s. 277 - 279
Hlavní autoři: Bayro, Allison, Jeong, Heejin
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 16.03.2024
Témata:
Affective computing
Computational modeling
Human-centered computing-Human computer interaction (HCI)-HCI design and evaluation methods-User models
Quantum computing
Solid modeling
Support vector machines
Three-dimensional displays
Virtual reality
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:This research explores the use of quantum machine learning-more specifically, the Quantum Support Vector Machine (QSVM)-to predict emotional reactions triggered in virtual reality (VR) settings. The study's experimental setup involved provoking different emotional states via VR gameplay elements and capturing the physiological signals of participants. These signals were paired with the subjects' self-reported emotions, measured through the Self-Assessment Manikin. The study then employed both conventional Support Vector Machines (SVM) and QSVM to classify the data. Notably, QSVM surpassed the traditional SVM in accurately predicting the levels of arousal and valence, achieving higher precision with a reduced set of features. This enhanced efficiency is likely due to QSVM's superior handling of the intricate patterns in emotional data and the quantum models' more effective computational resource usage. These results hold considerable potential for the field of affective computing within VR environments, indicating the advantageous prospects of quantum machine learning for the domain.
DOI:10.1109/VRW62533.2024.00054