Research on Chinese Micro-blog Sentiment Classification Based on Recurrent Neural Network

In order to improve the accuracy of Chinese micro-blog sentiment classification, a sentiment classification method based on a kind of improved recurrent neural network model was proposed. Firstly, for the extraction of the text intrinsic semantic features, a feature fusion method based on shallow and deep learning was adopted. Secondly, for the micro-blog text sentiment classification problem, an improved recurrent neural network model was adopted, which replaced the hidden layer of general recurrent neural network into LSTM structure. Finally, the classification accuracy rate was 85.04%, which was 3.17% higher than that of the traditional SVM with shallow learning features. The results confirm the effectiveness of the proposed method.