AUTHOR=Chen Yinchang , Dai Zhe TITLE=Mining of Movie Box Office and Movie Review Topics Using Social Network Big Data JOURNAL=Frontiers in Psychology VOLUME=Volume 13 - 2022 YEAR=2022 URL=https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2022.903380 DOI=10.3389/fpsyg.2022.903380 ISSN=1664-1078 ABSTRACT=In the film industry, there will inevitably be some high investment and low box office losses. In order to be able to solve this problem, movie box office and movie review topics are analyzed using social network big data. Firstly, the factors that affect the box office of the movie are analyzed. Secondly, continuous and discrete feature parts, text parts, and fusion parts are merged. The box office prediction model of mixed features using deep learning is established, and movie box office is predicted. Finally, compared with other algorithms and models, the box office prediction model of mixed features using deep learning is verified. Compared with Light Gradient Boosting Machine (LightGBM), when the number of iterations is 23 of the hybrid feature movie box office prediction model using deep learning, the Mean Square Error (MSE) value reaches the optimal value, which is 0.6625; when the number of iterations is 32, Mean Absolute Error (MAE) value reaches the optimal value, 0.1798. The proposed hybrid feature movie box office prediction model has higher prediction accuracy than LightGBM, and can achieve good prediction results. The designed model is used to predict the box office of 9 movies.