南京师范大学学报（工程技术版）2023，Vol.23Issue(4)：43-49,7.DOI:10.3969/j.issn.1672-1292.2023.04.006
基于改进BiLSTM算法的大学生心理健康问题研究分析
Research and Analysis on Psychological Health Problems of College Students Based on Improved BiLSTM Algorithm
摘要
Abstract
With the more and more wide application of deep learning models,the accuracy of the models continues to improve,providing feasibility for intelligent research and judgment systems.The psychological behavior of college students have both explicitness and implicitness.At present,implicit information is often overlooked in the process of psychological counseling.In order to extract implicit information more effectively,this paper uses the deep learning method to extract the psychological characteristics of college students'psychological interview data,and constructs an intelligent analysis algorithm for college students'psychological counseling data.In order to deepen the emotional orientation in the word vector,this paper uses the BERT model to replace the traditional Word2vec model.And the BiLSTM algorithm is used to strengthen the correlation between contexts.Experiments prove that the algorithm effectively obtains metaphorical and low-frequency semantic information in the process of psychological counseling,classifies psychological counseling data(positive emotion and negative emotion),and accurately warns the interview data of negative emotions.关键词
大学生心理/情感分析/深度学习/BiLSTM/词向量Key words
college student psychology/sentiment analysis/deep learning/BiLSTM/word vector分类
信息技术与安全科学引用本文复制引用
高星宇,施姣杰,陈坚..基于改进BiLSTM算法的大学生心理健康问题研究分析[J].南京师范大学学报（工程技术版）,2023,23(4):43-49,7.基金项目
浙江旅游职业学院招标课题专项项目(2023ZB02). （2023ZB02）
