摘要
针对传统的大学生心理健康状态评价方法存在计算复杂和准确率低的缺点,提出了一种基于数据降维和支持向量机的大学生心理健康状态评价方法。选择精神病性、偏执、敌对、恐怖、焦虑、抑郁、强迫症状、人际关系敏感和躯体化等9个维度的指标作为LLE-SVM模型的大学生心理健康状态评价模型的输入,将大学生心理健康状态作为LLE-SVM模型的输出,建立LLE-SVM的大学生心理健康状态评价模型。与SVM、BPNN和DT相比较,LLE-SVM能够有效提高大学生心理健康状态评价准确率。
In view of the traditional evaluation method of college students mental health status has the disadvantages of computational complexity and low accuracy,this paper presents a method for evaluating the mental health status of college students based on data dimension reduction and support vector machine.Nine dimensions of psychopathy,paranoid ideation,hostility,phobia,anxiety,depression,obsessive-compulsive symptoms,interpersonal sensitivity and somatization are selected as the input of LLE-SVM model.The LLE-SVM model is used to evaluate the mental health of college students.Compared with SVM,BPNN and DT,LLE-SVM can effectively improve the accuracy of mental health assessment of college students.
作者
李馥利
金敏
王雨佳
LI Fuli;JIN Min;WANG Yujia(School of Chemical Engineering and Modern Materials,Shangluo University,Shangluo 726000,China;School of Health Management,Shangluo University,Shangluo 726000,China;Xi’an Institute of Aeronautical Computing Technology,AVIC,Xi’an 710062,China)
出处
《微型电脑应用》
2021年第5期79-81,共3页
Microcomputer Applications
基金
商洛学院2020年学生工作研究课题(XSGZ2009)
2017年商洛学院教育教学改革研究项目(17jyjx129)。
关键词
支持向量机
心理健康状态评价
神经网络
局部线性嵌入算法
support vector machine
mental health status assessment
neural network
local linear embedding algorithm
