A Chinese Text Classification Model Based on XLNet and BiGRU

Aiming at the problems of insufficient Chinese text feature extraction and classification accuracy is not high, this paper further research on Chinese text classification based on deep learning. Comprehensive considering the different classification model, a Chinese text classification model is proposed, in which the initial features are extracted using the XLNet and BiGRU is used to vectorise the text information. At the same time, the attention mechanism is introduced into the BiGRU model to make it pay attention to the features of key words. Finally, a feature vector generated by this model is sent to the Softmax classifier for classification. By comparing the model with the other five text classification models in three datasets, the results show that the proposed text classification model has achieved good classification results in Chinese datasets, which verifies the accuracy and effectiveness of the model classification.