Application of Sentiment Classification of Weibo Comments Based on TextCNN Model
Zhongsheng Wang, Minlu Wang, Han Shen, Yonggui Han · 2022
The Internet has become the main medium for the dissemination of network public opinion. By analyzing the emotional development trend of emergencies, it is possible to explore the evolution of public opinion and identify potential risks, and to provide decision support for the guidance and control of public opinion. This paper combines Word2vec and network and text convolutional neural network (Convolutional Neural Network, TextCNN) model to achieve text sentiment classification for Weibo comments. Formalize the comment text, and segment the processed data through the stuttering word segmentation library, that is, extract the keyword words, and use the Word2vec tool to train the word vector for each word segmentation, and obtain the word embedding weight matrix. The embedding layer of the CNN model. In this paper, the comment text is normalized, and the processed data is segmented through Jieba, that is, the keyword words are extracted, and the Word2vec tool is used to train the word vector for each word segmentation, and the embedding of the word embedding weight matrix CNN model is obtained.