摘要
本文主要研究了GRU,LSTM等深度学习模型在谣言检测上的应用,判断微博文本是否为谣言类信息。考虑到新浪微博平台的图结构,一条微博文本对应着多条评论信息,评论中可能包含对该条文本的态度,例如赞成、反对、怀疑等。因此,本文在判断微博文本是否为谣言时,融合了评论信息,将评论看作一条时间线上的各个时刻,按照时间节点展开,作为时间序列模型每个时刻的输入,并且利用注意力机制衡量每个时间节点对最终语义表示的重要程度。实验结果表明,在加入评论信息及attention机制后,实验结果具有明显提升,最后达到92.66%的识别准确率。
This paper mainly studies the application of GRU and LSTM model in rumor detection research,to judge whether a microblogging text is rumor information or not.Considering the graph structure of sina microblogging,that is to say,a microblogging text corresponds to a number of comments,and comments may contain attitude towards this text,such as favor,opposition,doubt.This paper integrates comments information to rumor detection,regards comments as time nodes on a time line.After unfold by time,every comment can be treated as the input of time series model at every time.The paper also proposes attention mechanism,which weighs importance degree of every time node to final semantic representation.The experimental results show that,after adding comment information and attention mechanism,result improves significantly,finally reaches 92.66%.
作者
任文静
秦兵
刘挺
REN Wenjing;QIN Bing;LIU Ting(School of Computer Science and Technology,Harbin Institute of Technology,Harbin 150001,China)
出处
《智能计算机与应用》
2019年第3期300-303,共4页
Intelligent Computer and Applications
关键词
谣言检测
深度学习
新浪微博
分类
rumor detection
deep learning
sina microblogging
classification
