基于情感分析和Transformer模型的微博谣言检测  

Microblog Rumor Detection Based on Sentiment Analysis and Transformer Model

作　　者：冯茹嘉 张海军[1] 潘伟民[1] FENG Ru-jia;ZHANG Hai-jun;PAN Wei-min(College of Computer Science and Technology,Xinjiang Normal University,Urumqi 830054,China)

机构地区：[1]新疆师范大学计算机科学技术学院,新疆乌鲁木齐830054

出　　处：《计算机与现代化》2021年第10期1-7,共7页Computer and Modernization

基　　金：2019年度自治区创新环境(人才、基地)建设专项(人才专项计划——天山雪松计划)项目(2019XS08);国家自然科学基金-新疆联合基金重点项目(U1703261)。

摘　　要：针对微博文本以实现谣言检测为目标,深度挖掘微博正文内容的语义信息,并且着重强调用户在微博评论中体现的情感倾向性,提升谣言识别效果。为提高谣言检测的准确率,采取基于XLNet的词嵌入方法,使用Transformer的Encoder的模型提取微博正文内容的语义特征,并结合BiLSTM+Attention网络实现微博评论的情感特征的提取,将2种特征向量进行拼接融合,进一步丰富神经网络的输入特征,之后输出微博事件的分类结果,进而实现微博谣言检测。实验结果显示,该模型对谣言识别的正确率达到94.8%。Aiming at realizing the rumor detection on microblog,this paper deeply excavates the semantic information of the body content of microblog,and emphasizes the emotional tendency reflected by users in microblog comments,so as to improve the effect of rumor identification.In order to improve the rumor detection accuracy,based on XLNet word embedding method,the Transformer’s Encoder model is used to extract the semantic features of microblog body content.Combined with the BiLSTM+Attention network,the emotional feature extraction of microblog comments is realized.Two kinds of feature vectors are spliced and fused to further enrich the input features of neural network.Then,the microblog event classification results are output,and the microblog rumors detection is achieved.The experimental results show that the accuracy of the model in rumor recognition reaches 94.8%.

关 键 词：谣言检测 情感分析 XLNet Transformer模型 深度学习 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]