期刊論文
| 學年 | 113 |
|---|---|
| 學期 | 2 |
| 出版（發表）日期 | 2025-05-24 |
| 作品名稱 | Using Machine Learning Techniques to Discriminate Good and Poor Sleepers in Virtual Reality Environment |
| 作品名稱（其他語言） | |
| 著者 | En-Chen Chen; Tsai-Yen Li |
| 單位 | |
| 出版者 | |
| 著錄名稱、卷期、頁數 | Machine Learning and Soft Computing 2487, p.123-144 |
| 摘要 | Stressors in modern life have led to a concerning increase in insomnia, poor sleep quality, and nightmares, underscoring the urgent need for innovative solutions to alleviate the associated psychological and societal burdens. This study bridges neuroscience and technology by integrating electroencephalogram (EEG) and heart rate variability (HRV) data to evaluate the effectiveness of relaxation techniques under two distinct conditions: closed-eye relaxation and immersive virtual reality (VR) environments. These physiological markers were selected for their relevance in understanding brainwave activity and autonomic nervous system regulation, which play critical roles in sleep quality and emotional well-being. Employing Few-shot learning approaches combined with machine learning algorithms such as k-nearest Neighbors (KNN) and Support Vector Machine (SVM), this study sought to address the common limitation of small sample sizes in clinical research. Results revealed that SVM consistently outperformed KNN in classification accuracy and F1 scores. Furthermore, cross-validation and principal component analysis (PCA) demonstrated that VR-based relaxation techniques significantly enhance the ability to differentiate between individuals with good and poor sleep. The study emphasizes the critical interplay between brainwave adaptability and autonomic nervous system regulation, revealing that even without external stimuli, individuals with poor sleep are prone to generating complex emotional and anxious responses through cognitive and attentional processes. This highlights the importance of targeted interventions for improving sleep quality. This research contributes to the scientific community by presenting a novel, interdisciplinary approach that combines VR technology with physiological monitoring and machine learning. It offers insights into the potential of VR-based therapies in advancing personalized treatment for insomnia and poor sleep. |
| 關鍵字 | |
| 語言 | en |
| ISSN | 1865-0929 |
| 期刊性質 | 國外 |
| 收錄於 | EI |
| 產學合作 | |
| 通訊作者 | |
| 審稿制度 | 否 |
| 國別 | JPN |
| 公開徵稿 | |
| 出版型式 | ,電子版 |
| 相關連結 |
機構典藏連結 ( http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/129693 ) |