摘要
为了更智能、准确地从高校学生的社交动态中分析学生们的心理健康状态,该研究提出了一种基于多模态社交情感分类的高校学生心理健康分析方法。针对高校学生群体情感表现的复杂性,提出了一种将情感状态表达划分主体情感和侧面情感的分析方法;针对社交动态数据模态的多样性,提出一种多模态数据融合方法。实验结果表明,论文提出的多模态社交情感分类方法在构建的高校学生社交动态数据集上主要情感分类得的准确率达到89.8%,并在多个公开数据集上相对于基准算法提高了4%~6%的分类准确率。
To analyze the mental health of college students from the social feeds of them more intelligently and efficiently,this paper proposes a multimodal social emotion classification based mental health analysis method. Considering the complexity of college students’ emotion expression,it divides emotional state into primary emotion and secondary emotion. Aiming at the diverse modals of social feeds data,a multi-modal data fusion method is proposed is proposed. Experimental results show that the proposed multimodal social emotion classification method achieves an accuracy of 89.8% on the collected college students’ social feeds data set,and improves the classification accuracy of 4%~6% compared with the benchmark algorithm on multiple public data sets.
作者
王芳
赵小明
WANG Fang;ZHAO Xiaoming(College of Computer Science and Technology,Qingdao Institute of Software,China University of Petroleum,Qingdao 266580;College of Control Science and Engineering,China University of Petroleum,Qingdao 266580)
出处
《计算机与数字工程》
2022年第10期2166-2170,共5页
Computer & Digital Engineering
基金
国家自然科学基金项目(编号:62072469)
2021年度教育部人文社会科学研究专项任务项目(编号:21JDSZ3201)资助。
关键词
深度学习
情感分类
心理健康
多模态融合
deep learning
emotion classification
mental health
multimodal fusion
