隨著網際網路的普及和上網人口的增加，對於網路成癮所帶來的工作、家庭、學業和身心健康等問題關注日益增加。網路成癮的判定方式為透過問卷填寫，並由心理專業人士進行評估。然而，這種方式有著需要大量人力的限制，以及調查難以普及等問題。這將造成潛在成癮者無法被發現，並得到適當的輔導，導致其身心健康持續受損。網路成癮會因為情緒低落、睡眠障礙等因素，透過自主神經系統，持續影響成癮者的身心健康，造成睡眠品質下降和精神疲憊等現象。本研究運用穿戴式裝置收集大學生的心率數據，以及網路成癮問卷調查結果，判別受試者的成癮狀態。利用深度學習的方法，學習網路成癮與非成癮者的心率特徵，建立一個基於心率對網路成癮狀態預測的模型。研究的結果能夠提供心理專業人員作為篩選潛在網路成癮者的方法，協助其發現網路成癮的病患，減少潛在成癮者未能接受輔導以至於身心健康持續受損的問題。 As the Internet becomes more popular and the population of Internet users increases, there is a growing concern about the work, family, academic, and physical and mental health problems associated with Internet addiction. Internet addiction is determined by completing a questionnaire and having it assessed by a psychological professional. However, this approach is limited by the large amount of manpower required and the difficulty of generalizing the survey. This will result in potential addicts not being identified and receiving appropriate counseling, resulting in continued damage to their physical and mental health. Internet addiction can continue to affect the physical and mental health of addicts through the autonomic nervous system due to factors such as depression and Sleep Disorders, resulting in decreased sleep quality and mental fatigue.This study used a wearable device to collect heart rate data from university students and Internet addiction questionnaires to identify the addiction status of the subjects. A Deep Learning approach was used to learn the heart rate characteristics of Internet addicts and non-addicts, and a model was developed to predict Internet addiction based on heart rate. The results of the study can be used by psychological professionals as a way to screen for potential Internet addicts, to help them identify patients with Internet addiction, and to reduce the problem of potential addicts who are not receiving counseling and whose physical and mental health continues to suffer.
