基于语义增强的在线健康社区情感分析研究  

Sentiment Analysis of Online Healthy Community Based on Semantic Enhancement

作　　者：韩普 叶东宇 HAN Pu;YE Dongyu(School of Management,Nanjing University of Posts&Telecommunications,Nanjing 210003;Jiangsu Provincial Key Laboratory of Data Engineering and Knowledge Service,Nanjing 210023)

机构地区：[1]南京邮电大学管理学院,南京210003 [2]江苏省数据工程与知识服务重点实验室,南京210023

出　　处：《科技情报研究》2024年第2期88-99,共12页Scientific Information Research

基　　金：国家社会科学基金项目“面向多模态医疗健康数据的知识组织模式研究”(编号:22BTQ096);江苏高校青蓝工程和南京邮电大学1311人才计划资助;江苏省研究生科研创新计划资助(编号:KYCX22_0870)。

摘　　要：[目的/意义]为了更充分利用文本依存句法信息和先验情感知识在情感分析中的价值,提出了一种语义增强的在线健康社区情感分析模型。[方法/过程]首先预处理在线健康社区数据,并通过BERT生成特征向量;接着基于双通道思想,利用TextCNN和BiLSTM分别抽取在线评论文本的局部和全局信息,然后在GAT中融入情感知识和文本依存句法信息进行语义增强;最后进行双通道特征拼接,并在全连接层实现在线健康社区情感极性判断。[结果/结论]通过对31 718条在线健康社区评论数据进行对照实验发现,基于语义增强的BERT-TBGH模型准确率达到90.77%,相比基准模型TextCNN和BiLSTM分别提升了10.57%和7.79%,引入情感知识和字粒度依存句法信息后,准确率分别提升了1.85%和1.00%。文章提出的基于语义增强的BERTTBGH模型能够有效提升在线健康社区情感分析效果。[Purpose/significance]In order to make full use of the value of text dependent syntactic information and prior emotion knowledge in emotion analysis,a semantic enhanced online healthy community emotion analysis model was proposed.[Method/process]Firstly,feature vectors for pre-processed online health community data are generated by Word2Vec and BERT;then local and global information of online review text are extracted using TextCNN and BiLSTM respectively based on dual-channel idea;then sentiment knowledge and dependency grammar information are merged in graph attention networks for semantic enhancement;finally,dual-channel features are fused and perform online health community sentiment classification in fully connected layer.[Result/conclusion]The comparative experiments on 31718 online health community comments show that the accuracy of the BERT-TBGH model based on semantic enhancement reaches 90.77%,which is 10.57%and 7.79%higher than the classical models TextCNN and BiLSTM and is 1.85%and 1.00%higher after introducing sentiment knowledge and character-level dependency syntactic information.The proposed model based on semantic enhanced BERT-TBGH model can effectively improve the effect of online healthy community emotion analysis.The defect of this article is that the experiment was limited to the online health community dataset and was not further validated on larger datasets.

关 键 词：情感分析 依存句法分析 图神经网络 语义增强 BERT-TBGH 在线健康社区 

分 类 号：G358[文化科学—情报学] TP391[自动化与计算机技术—计算机应用技术]