基于改进的卷积记忆神经网络的文本情感分类

Text Sentiment Classification Based on Improved Convolutional Memory Neural Network

针对如何将评论文本按照不同的情感进行分类的问题,提出改进的卷积记忆神经网络模型。该模型融合了卷积神经网络和双向长短期记忆神经网络在信息提取上的优势,并在卷积神经网络中进行动态池化处理,从而提取更多的显著文本特征。实验结果表明,改进卷积记忆神经网络的准确率、精度、召回率、F1测度分别达到92.41%、92.32%、93.27%、93.63%。相比于卷积神经网络、双向长短期记忆神经网络、卷积记忆神经网络,该模型在处理和评论文本情感分类问题上具有较好的效果。

An improved convolutional memory neural network model is proposed for how to classify comment texts according to different emotions.The model combines the advantages of convolutional neural networks and bidirectional long-and short-term memory neural networks in information extraction,and makes dynamic pooling processing in convolutional neural networks to extract more significant text features.The experimental results show that the accuracy,precision,recall and F1-measure of improved convolutional memory neural network are 92.41%,92.32%,93.27%,93.63%,respectively.Compared with convolutional neural networks,bidirectional long-term and short-term memory neural network,and convolutional memory neural networks,this model has a good effect in dealing with the emotional classification of comment texts.