基于LERT和双通道模型的微博评论情感分析研究

Research on sentiment analysis of microblog comments based on LERT and dual channel model

为了高效监管舆论环境、加强网络舆情引导,提出一种基于LERT和双通道模型的微博评论情感分析方法。使用LERT预训练语言模型进行语义分析,通过BiGRU和TextCNN双通道结构获取文本信息,将双通道处理结果输入分类层获得情感分析结果。实验表明,该模型方法的正确率和F1值分别达到92.09%和94.38%,优于其他深度学习模型。

In order to efficiently supervise public opinion environment and strengthen online public opinion guidance,a sentiment analysis model of microblog comments based on LERT and dual channel model is proposed.Firstly,the LERT pre-trained language model is used for semantic analysis.Then the text information is obtained through BiGRU and TextCNN dual channel structure.Finally the dual channel processing results are input into the classification layer to obtain the sentiment analysis results.Experiments show that the accuracy and F1 value of the proposed model reach 92.09%and 94.38%,respectively,which are superior to other deep learning models.