福建师范大学学报（自然科学版）2025，Vol.41Issue(6)：45-53,9.DOI:10.12046/j.issn.1000-5277.2024100030
基于深度学习的高校学生微博情感分析
Sentiment Analysis of Weibo Posts by College Students Based on Deep Learning
摘要
Abstract
This study employs deep learning techniques to analyze Weibo posts by college students in order to identify feedback information,thereby enabling insight into their emotional dy-namics in real time.It maps words from the posts into a vector space using Word2vec,providing rich input features for the long short-term memory(LSTM)and Bidirectional long short-term memory(Bi-LSTM)models.These two deep learning models are then used for sentiment prediction.Com-parative experiments between LSTM and Bi-LSTM models with different network layers demonstrate that the Bi-LSTM model,when configured with two network layers,achieves the best accuracy and F1 score in sentiment classification tasks.关键词
深度学习/情感分析/微博/长短期记忆网络/双向长短期记忆Key words
deep learning/sentiment analysis/Weibo/LSTM/Bi-LSTM分类
信息技术与安全科学引用本文复制引用
张舒页,林兵..基于深度学习的高校学生微博情感分析[J].福建师范大学学报（自然科学版）,2025,41(6):45-53,9.基金项目
国家自然科学基金项目(62072108) （62072108）
福建省高校物理学学科联盟教学改革项目(FJPHYS-2022-B02) （FJPHYS-2022-B02）
福建省科技经济融合服务平台项目(2023XRH001) （2023XRH001）
福厦泉国家自主创新示范区协同创新平台项目(2022FX5) （2022FX5）
福建省高校产学合作资助项目(2022H6024,2021H6026) （2022H6024,2021H6026）
