Sentiment Analysis Method of Microblog Text Based on Transformer
In this paper,the self-attention mechanism in the Transformer model is used to simulate the human brain nervous system to extract the features of the microblog text,and the TextCNN layer is used to convolve the word vectors passed through the Transformer to obtain the timing information between adjacent word vectors.The activation function optimizes the model by Tanh,and finally the temporal attention weights obtained by the convolutional layers are applied to text classification.The experimental re-sults show that the accuracy of the model proposed in this paper on the NLP&CC2013 dataset is improved by 0.38%compared with the Transformer model,and the precision,recall and F1 value are improved to a certain extent.
万方数据
维普
