AI赋能心理健康服务——基于LDA模型的热点识别与发展导向研究

AI-Empowered Mental Health Services—Hotspot Identification and Development Orientation Based on LDA Model

摘要: 该研究旨在通过量化方法，系统梳理“人工智能赋能心理健康服务”领域的研究热点和发展趋势。研究者以中国知网(CNKI)收录的409篇相关文献的摘要和关键词为语料，采用LDA (Latent Dirichlet Allocation)主题模型进行文本挖掘。通过计算困惑度确定最优主题数为11，并进一步计算主题强度，识别出五个核心热点主题：1) AI焦虑识别与辅导干预；2) AI心理咨询与跨群体应用；3) 高校AI心理教育趋势；4) AI赋能辅导员心理育人；5) AI情感伦理。研究发现，这五个主题呈现出从“技术识别”到“服务应用”再到“伦理反思”的清晰递进结构。作者认为，这表明该领域的研究已从单纯的技术可行性探索，逐步转向服务模式构建与人文价值的审视，并对未来的研究方向提出了具体建议。

Abstract: This study aims to systematically identify research hotspots and development trends in the field of "AI-empowered mental health services" through quantitative methods. A corpus comprising the abstracts and keywords of 409 relevant articles indexed in the China National Knowledge Infrastructure (CNKI) was constructed. The Latent Dirichlet Allocation (LDA) topic model was employed for text mining. By calculating perplexity, the optimal number of topics was determined as 11. Topic strengths were further computed, leading to the identification of five core research hotspots: (1) AI-assisted anxiety recognition and intervention, (2) AI-based psychological counseling and cross-population applications, (3) trends in AI mental health education in universities, (4) AI empowerment of counselors in psychological education, and (5) AI emotional ethics. The study found that these five topics exhibit a clear progressive structure—from technical identification, through service application, to ethical reflection. The authors argue that this indicates the field has gradually shifted from exploring purely technical feasibility toward constructing service models and examining humanistic values. Concrete suggestions for future research directions are also proposed.