Chinese Text Sentiment Classification Model Based on FastText and Multi-scale Deep Pyramid Convolutional Neural Network

In order to improve the efficiency of Chinese text sentiment classification, a text sentiment classification model based on FastText and multi-scale DPCNN are proposed. First, a text vector matrix is constructed using the FastText model. Then, multiple feature maps are extracted from the text vector matrix with a multi-scale filter. Finally, multiple feature maps are fused and input into the DPCNN model for sentiment classification. Experiments are carried out on the Chinese sentiment mining corpus (ChnSentiCorp) dataset, and the comparison results of multiple groups of experiments show that the proposed model can effectively improve the accuracy of text sentiment classification compared with other algorithms.