首页|融合LDA-LSTM算法的微博档案关注度和情感分析

融合LDA-LSTM算法的微博档案关注度和情感分析

扫码查看
为了解析《档案法》实施三年后社会公众对档案的关注度与情感态度,利用Python工具和LDA模型对数据进行提取和主题聚类,获得不同时间下档案热点主题;采用LSTM模型得到各档案主题的情感倾向,分析用户产生不同情感倾向的原因.根据各主题间的联系,得到档案项目、影视娱乐档案、学生-学校档案以及专项档案四类主题.各类档案主题具有较高的积极倾向,说明公众对档案事业的发展比较支持和理解,对于消极倾向较高的学生-学校档案类别,有关部门应加强档案宣传教育,从根本上保障公众利用档案的权利,增强全社会的档案意识.
Analysis of emotion and attention of Weibo archives based on LDA-LSTM algorithm
In order to analyze the public's attention and emotional attitude towards the archives two years after the implemen-tation of the Archives Law,the authors used Python tools and LDA model to extract and cluster the data,hot topics of archives at different times were obtained in this paper;the LSTM model was utilized to obtain the emotional tendency of each file theme and analyzed the reasons for the different emotional tendencies of users.According to the connection between themes,there were four categories of archives projects,such as film and television entertainment files,student-school archives and special archives;all kinds of archives categories had high positive tendency,which indicating the public support and understand the development of ar-chives.For the students with high negative tendency-school archives category,the relevant departments should strengthen the pub-licity and education of archives,protect the public's right to use archives fundamentally,and enhance the archives awareness of the whole society.

attention of archivesLDA-LSTM algorithmemotion analysisSina Weibo

孙思怡、王家强、罗子江

展开 >

贵阳人文科技学院经济与管理学院,贵阳 550025

贵州财经大学信息学院,贵阳 550025

顺德职业技术学院智能制造学院,顺德 528333

档案关注度 LDA-LSTM算法 情感分析 新浪微博

2024

现代计算机
中大控股

现代计算机

影响因子:0.292
ISSN:1007-1423
年,卷(期):2024.30(7)
  • 12