A Review of Early Box Office Prediction Based on Social Media

Abstract

Abstract. A wide range of box office prediction surveys based on artificial intelligence have made progress. However, early box office prediction remains imperfect due to significant uncertainty before a film's release, leading to increased interest in this area of research. Despite this, research is still in its early stages. To address this issue, we review early box office prediction methods based on social media data. By examining three approaches, we derive a generic process for early box office prediction. This paper compares the advantages and disadvantages of machine learning- and deep learning-based prediction methods. Additionally, we analyze the research trends and challenges in the field, providing a reference for researchers who are new to this area.