Research on Chinese Movie Reviews Based on Latent Dirichlet Allocation Topic Model

With the rapid development of the Internet, information acquisition has become more accessible, and a large number of online movie reviews cover the characteristics of audience preferences in the movie consumption market. Taking the top 10 popular movies in the Chinese box office as an example, this paper collects hot reviews, counts word frequency and visualizes the word cloud to show the keywords. At the same time, this paper uses the topic model to summarize key topics, so that we can gain insight into the hot spots. Finally, we conclude that the audience's attention mainly focus on the movie's theme, plot content, scene design, editing technology, character shaping, etc. Directors and actors have received a lot of attention, and the attention of domestic movies has increased significantly.