现代情报2025，Vol.45Issue(12)：118-129,12.DOI:10.3969/j.issn.1008-0821.2025.12.010
网络舆情事件用户在线评论行为预测研究
Research on Predicting Users'Online Commenting Behavior in Online Public Opinion Events
摘要
Abstract
[Purpose/Significance]In the context of increasingly complex online public opinion,a DRSN-CW-LSTM based model for predicting user comment behavior is proposed to provide reference for network public opinion governance,crisis response,and public opinion guidance strategy formulation.[Method/Process]This model comprehensively consi-ders and extracts different user features and public opinion event features to predict user comment behavior.It integrates deep residual shrinkage network and long short-term memory network,and introduces soft thresholding as a nonlinear layer into the ResNet structure to achieve better feature learning performance.[Result/Conclusion]Experimental results on real-world datasets of online public opinion events show that,compared with other baseline models,the proposed method is highly practical and more accurate.关键词
用户评论行为/深度残差收缩网络/DRSN-CW-LSTM网络/网络舆情事件/评论行为预测Key words
user comment behavior/deep residual shrinkage network/DRSN-CW-LSTM Network/online public opinion events/prediction of comment behavior引用本文复制引用
沈旺,李欣,孙珂,李昕娱..网络舆情事件用户在线评论行为预测研究[J].现代情报,2025,45(12):118-129,12.基金项目
国家自然科学基金青年科学基金项目"基于复杂系统层次演化的社交媒体群体行为演化机制及特殊行为群体发现研究"(项目编号:72304111). （项目编号:72304111）
