%0 Journal Article
%T 随机森林模型在预测大一新生智能手机成瘾中的应用<br>Application of the Random Forest Model in Predicting Smartphone Addiction among First-Year College Students
%A 毛心格
%J Advances in Psychology
%P 30-37
%@ 2160-7281
%D 2024
%I Hans Publishing
%R 10.12677/ap.2024.1410694
%X 目的&#65306;应用随机森林算法预测大一新生智能手机成瘾类别的效果&#65292;并分析影响大一新生智能手机成瘾的重要因素。方法&#65306;收集某高校2530名大一新生的心理学和人口学信息&#65292;采用随机森林构建判断大一新生智能手机成瘾程度的三分类模型&#65292;选择AUC值、F1值、召回率和精准率作为模型评价指标。结果&#65306;随机森林的分类效果良好&#65292;同时通过影响因素的重要性分析发现&#65292;预测智能手机成瘾的最重要的五个因素分别是学业倦怠、错失焦虑、自我控制、拖延行为和社交焦虑。结论&#65306;随机森林模型能够有效预测大一新生的智能手机成瘾。<br>Objective: This paper aims to evaluate the effectiveness of applying the Random Forest algorithm in predicting the categories of smartphone addiction among first-year college students, and to analyze the key factors influencing smartphone addiction in this population. Methods: Psychological and demographic data were collected from 2,530 first-year students at a university. A three-class classification model was constructed using the Random Forest algorithm to determine the level of smartphone addiction among these students. The model&#8217;s performance was evaluated using AUC, F1-score, recall, and precision as metrics. Results: The Random Forest model demonstrated good classification performance. Furthermore, an analysis of the importance of influencing factors revealed that the five most significant predictors of smartphone addiction were academic burnout, fear of missing out, self-control, procrastination behavior, and social anxiety. Conclusion: The Random Forest model is effective in predicting smartphone addiction among first-year college students.
%K 智能手机成瘾&#65292
%K 随机森林&#65292
%K 大学生&#65292
%K 自我控制&#65292
%K 学业倦怠<br>Smartphone Addiction
%K Random Forest
%K College Students
%K Self-Control
%K Academic Burnout
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=97809