Research on Social Network Sentiment Analysis Based on Big Data

With the development of the Internet, users generate a large amount of text content on the platform, reflecting their emotional states, opinions and life experiences, and sentiment analysis, as an important tool to understand users’ emotional tendencies, has attracted more and more attention. However, sentiment analysis in social networks faces challenges such as high data dimension, complex structure and emotion expression, and traditional methods are difficult to accurately identify subtle emotional differences in complex data sets. This study proposed an innovative sentiment analysis model. Based on the data of Weibo platform, it collected the text content, publication time, number of retweets and number of comments posted by users through API interface, and imported 1,453,051 microblog data into an efficient distributed computing framework to extract deep emotion features. Identify emotions in text based on deep learning algorithms and sentiment dictionaries. The experimental results of this study show that the model performs well in emotion classification tasks, and the accuracy of the convolutional neural network model reaches 90.5%, which is superior to support vector machines and long short-term memory networks.