基于注意力机制的微博文本情感分类模型  

Research on Sentiment Classification of Weibo Text Based on Attention Mechanism

作　　者：师贝贝 Shi Beibei(Beijing Key Laboratory of Network Culture and Digital Communication,Beijing Information Science and Technology University,Beijing 100101,China)

机构地区：[1]北京信息科技大学网络文化与数字传播北京市重点实验室,北京100101

出　　处：《信息与电脑》2020年第19期52-54,共3页Information & Computer

摘　　要：随着信息时代的到来,微博作为数据共享和信息服务的载体,已成为大众化服务模式下意见发布和情感反馈的平台。为了更好地研究微博文本情感分析,本文以biGRU为基础,提出了一种基于注意机制的文本情感分类模型。通过实验研究,将该模型与其他情感分类模型进行情感分析对比,验证该模型的优越性。以“微博”抓取的实际数据作为实验数据,利用训练好的模型对微博文本进行情感分析。实验结果表明:与传统的基于SVM的文本情感分类方法相比,基于注意力机制的微博文本情感分类方法的准确率有了明显提高。With the advent of the information open era,as a carrier of data sharing and information services,Weibo has become a platform for user information,opinion release and emotional feedback under the popular service model.In order to better study the sentiment analysis of Weibo texts,this paper proposes a text sentiment classification model based on attention mechanism based on biGRU.Through experimental research,the model is compared with other sentiment classification models for sentiment analysis to verify the superiority of this model.The actual data captured by"Weibo"is used as experimental data,different classifiers are trained,and sentiment analysis is performed on the Weibo text using the trained model.Experimental results show that compared with the traditional SVM-based text sentiment classification method,the accuracy of the microblog text sentiment classification method based on the attention mechanism has been greatly improved.

关 键 词：注意力机制 biGRU 微博文本 情感分析 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]