【作 者】：
周炫余， 刘 林， 陈圆圆， 洪嘉玲， 卢 笑
【关 键 词 】：
心理健康教育； 自动评估； 多模态融合计算； 深度学习； 生态瞬时评估
【栏 目】：
课程与教学
【中文摘要】：
快速准确地评估大学生心理健康状况是高校心理健康教育的重要任务，也是高校心理工作实现精准干预和提供个性化教育服务的基础。传统评估方法存在评估实时性不高、单一模态数据评估效果差、社会称许性反应偏误等问题。研究基于生态瞬时评估理论，以深度学习算法为手段，提出了一种基于多模态数据融合计算的大学生心理健康自动评估方法。该方法在自构建的多模态心理评估数据集（JA-IPAD）上测试表明：该模型能够精准评估大学生的心理健康状态，在智慧学习环境中具有良好的应用前景，能为完善学生心理档案、精准干预学生心理、优化心理健康服务提供决策依据和技术支撑，也能为促进高校智慧化心理健康教育作出贡献。
【英文摘要】：
Rapid and accurate assessment of college students' mental health is an important task of mental health education in colleges and universities. It is also the basis for accurate intervention and personalized education services in the psychological work of colleges and universities. Traditional assessment methods have some problems, such as low real-time assessment, poor assessment effect of single-modal data, and bias of social desirable responses. Based on the ecological transient assessment theory, this paper proposes an automatic mental health assessment method for college students based on multimodal data fusion calculation by means of deep learning algorithms.This method is tested on a self-constructed multimodal psychological assessment dataset (JA-IPAD). The results show that this model can accurately assess the mental health status of college students, and has a good application prospect in the intelligent learning environment. It can provide decision-making basis and technical support for improving students' psychological files, accurately intervening students' psychology and optimizing mental health services, and also contribute to promoting intelligent mental health education in colleges and universities.