多尺度卷积循环神经网络的情感分类技术

结合卷积神经网络对于特征提取的优势和循环神经网络的长短时记忆算法的优势,提出一种新的基于多尺度的卷积循环神经网络模型,利用卷积神经网络中的多尺寸滤波器提取出具有丰富上下文关系的词特征,循环神经网络中的长短时记忆算法将提取到的词特征与句子的结构联系起来,从而完成文本情感分类任务.实验结果表明:与多种文本情感分类方法相比,文中算法具有较高的精度.

Sentiment Classification With Multiscale Convolutional Recurrent Neural Network

Combining the advantages of convolution neural network(CNN)for feature extraction and recurrent neural network(RNN)for long shot-term memory, a new model based on multiscale convolutional recurrent neural network is proposed. This model utilize multi-size filter of CNN to extract word feature which contain a rich context information and use the long short-term memory algorithm of RNN to reflect the grammatical relations about the word and the sentence, and then completing the sentiment classification task. The experimental results show that: through comparing with many other sentiment classification, this new model has a high accuracy.