【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。更多还原

【Abstract】 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.更多还原