A comparative analysis of Machine Learning approaches for Movie Success Prediction

The success of a movie is very critical not only for the producers of the movie but also for the thousands of workers who work behind the scenes. Their livelihood depends on the income produced from the movie. The accurate prediction of whether a movie will be a hit, or a flop is a non-trivial task since it involves many unknown parameters. In this regard, the application of machine learning (ML) in predicting whether a movie will be successful, or flop can go a long way in reducing the financial burden of all the stakeholders. In this paper, the focus is on developing an application which will help in predicting the success of the movie at early stages which will help investors to invest in movie and analysis is performed on some of the patterns from the IMDb dataset. By collecting data from a lot of sources and applying various machine learning techniques, the system predicts an approximate success rate of a movie based on its success by examining past statistics from sources such as IMDb, Rotten Tomato. Experimental results demonstrate that the score in the testing phase is very impressive. The further paper also concludes predicting which set of actors or actresses are best to have the maximum profit of the movie. This exploration emphasizes the significance of prediction in the business industry. As all investments are based on these types of prediction only.