基于LSTM网络的大学生心理健康特征提取方法  

Feature extraction method of College Students’mental health based on LSTM network

作　　者：李平[1] LI Ping(Psychological Center of Student Office of Yan’an Vocational and Technical College,Yan’an 716000,Shaanxi Province,China)

机构地区：[1]延安职业技术学院学生处心理中心,陕西延安716000

出　　处：《信息技术》2023年第7期131-135,共5页Information Technology

摘　　要：为了提高大学生心理健康特征提取精度,提出了基于长短期记忆网络(Long Short-Term Memory,LSTM)的大学生心理健康特征提取方法。在建立大学生心理健康特征提取模型的基础上,利用Word2vec词向量训练心理健康语料库数据,并将其作为模型的原始特征向量,采用LSTM网络提取数据特征,获取大学生心理健康特征分类模型,完成大学生心理健康特征提取。仿真结果表明,提出方法可以准确提取大学生心理健康特征的同时,降低了大学生心理健康特征提取的时间。In order to improve the accuracy of College Students’mental health feature extraction,a college students’mental health feature extraction method based on Long Short-Term Memory(LSTM)network is proposed.Based on the establishment of College Students’mental health feature extraction model,the Word2vec word vector is used to train the mental health corpus data,which is used as the original feature vector of the model.The LSTM network is used to extract the data features,obtain the college students’mental health feature classification model,and complete the college students’mental health feature extraction.Simulation results show that the proposed method can accurately extract college students’mental health features and reduce the extraction time of College Students’mental health features.

关 键 词：长短期记忆网络 Word2vec模型 心理健康语料库 心理健康特征 特征分类 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]