Research on Emotional Tendency Analysis of Weibo Comments Based on Deep Neural Network

Abstract

With the rapid development of China's economy, social media has also developed rapidly, among which Weibo has grown into a more popular social media. Users will comment on hot events on the Weibo platform. Therefore, how to conduct sentiment analysis on user comments has become necessary research for relevant departments. Through natural processing technology, practical information can be extracted from the text of Weibo, which can provide appropriate decision-making for China's network security monitoring, prediction of potential problems, and product analysis. This paper proposes a method based on Word2Vec and a recurrent neural network improved for long and short-term memory networks to conduct sentiment analysis of Chinese Weibo comments through the integration of deep learning. And through experiments, the accuracy of sentiment analysis of Weibo comments in this article has reached 92%. Through the research of this article, we can better guide and supervise the sentiment tendency of public opinion for relevant departments.