摘要
【目的】为进一步挖掘突发公共卫生事件微博文本深层语义信息,提出一种基于特征融合和注意力机制的多通道微博情感分析模型。【方法】首先,在特征向量嵌入层利用Word2Vec和FastText生成词向量,并与词性特征向量和位置特征向量进行融合;其次,基于CNN和BiLSTM构建多通道层以提取微博文本局部和全局特征;接着,通过构建注意力机制层以提取微博文本重要语义特征;最后,在融合层合并多通道输出结果,并在输出层采用Softmax函数进行情感分类。【结果】在42384条突发公共卫生事件新冠疫情微博数据上进行对照实验,结果表明所提情感分析模型F1值达到90.21%,较基准模型CNN和BiLSTM分别提升9.71个百分点和9.14个百分点。【局限】所构建的数据集规模较小,并且尚未考虑图片和语音等多模态信息。【结论】所提模型在深度学习和多通道基础上,通过引入注意力机制并融合CNN和BiLSTM捕获的微博文本局部和全局语义特征达到了最优效果,进一步推动了微博情感分析研究进展。
[Objective]This paper proposes a multi-channel MCMF-A model for Weibo posts based on feature fusion and attention mechanism,aiming to further explore the semantic information of public health emergency.[Methods]Firstly,we generated word vectors with Word2 vec and FastText at the feature vector embedding level,which were merged with the vectors of part-of-speech features and position features.Secondly,we constructed multi-channel layer based on CNN and BiLSTM to extract local and global features of Weibo posts.Thirdly,we utilized the attention mechanism to extract important features of the texts.Finally,we merged the multi-channel output results,and used the softmax function for sentiment classification.[Results]We examined MCMF-A model with 42384 Weibo posts on COVID-19.The F1 value of the proposed model reached 90.21%,which was 9.71%and 9.14%higher than the benchmark CNN and BiLSTM models.[Limitations]More research is needed to expand the experiment data size to include more small and multi-modal information such as images and voices.[Conclusions]The proposed model could effectively conduct sentiment analysis with Weibo posts.
作者
韩普
张伟
张展鹏
王宇欣
方浩宇
Han Pu;Zhang Wei;Zhang Zhanpeng;Wang Yuxin;Fang Haoyu(School of Management,Nanjing University of Posts&Telecommunications,Nanjing 210003,China;Jiangsu Provincial Key Laboratory of Data Engineering and Knowledge Service,Nanjing 210023,China)
出处
《数据分析与知识发现》
CSSCI
CSCD
北大核心
2021年第11期68-79,共12页
Data Analysis and Knowledge Discovery
基金
国家社会科学基金项目(项目编号:17CTQ022)
国家级大学生创新训练计划项目(项目编号:SZDG2020040)
江苏研究生科研创新计划基金项目(项目编号:KYCX20_0844)的研究成果之一。
关键词
多通道
特征融合
深度学习
情感分析
突发公共卫生事件
Multi-Channel
Feature Fusion
Deep Learning
Sentiment Analysis
Public Health Emergencies
