基于CNN-LSTM电影评论的情感分析  

Emotional Analysis of Movie Review Based on CNN-LSTM

作　　者：孙敏 李旸[1] 余大为 张恩宝 李倩倩 SUN Min;LI Yang;YU Dawei;ZHANG Enbao;LI Qianqian(Anhui Agriculture University,Hefei 230036,China)

机构地区：[1]安徽农业大学信息与计算机学院

出　　处：《洛阳理工学院学报（自然科学版）》2019年第4期71-77,共7页Journal of Luoyang Institute of Science and Technology：Natural Science Edition

摘　　要：情感性评论能反映出人们的意见信息,对这些文本进行情感分析可以获得许多有价值的信息。将CNN和LSTM相结合,对公开的电影评论数据集IMDB分别采用CNN、LSTM和CNN-LSTM模型进行情感分析和研究比较。实验结果表明,CNN-LSTM网络模型比单个的CNN网络模型和单个LSTM网络模型具有更高的精度,可以解决这种短文本中的情感分析问题。Due to the confusion about the application of different emotional analysis models of the movie review data, in this study, CNN, LSTM and CNN-LSTM models are used respectively to analyze open IMDB, then a comparative study on their performances is carried out. The experimental results show that CNN-LSTM network model has higher prediction accuracy than single CNN network model and single LSTM network model. It can solve the problem of emotional analysis in such short texts.

关 键 词：情感分析 电影评论 CNN-LSTM 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]