A Comparative Study of Hollywood Movie Successfulness Prediction Model

The movie industry is a highly competitive industry with a lot of new movies are queued to be released each year. Movie making is subject to potential profits or loss in the magnitude of billion of dollars making this industry very risky. Predicting the successfulness of movie based on its financial performance prior to the release date is valuable in order to reduce number of uncertainties faced by decision makers such as producers, distributors, and exhibitors. Using the concept of machine learning, we suggest a classification model to predict the successfulness of a Hollywood Movie using Artificial Neural Network, Naïve Bayes and Support Vector Machine. The objective of this research is to compare classification algorithms performance for predicting the successfulness of Hollywood movies before they are being released. Artificial Neural Network produces the best model in terms of performance in predicting a movie successfulness. Reaching 80% of accuracy and having above 80% of F-measure, precision, and recall suggests that Artificial Neural Network is a good model to assist producers, distributors and exhibitors assess risks.