PREDICTING MOVIE BOX OFFICE REVENUE USING ARTIFICIAL NEURAL NETWORK

• Main Author: Marsha, Crystal
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/49383
• Institution: Institut Teknologi Bandung

The film industry has grown immensely over the past few decades generating billions of dollars of revenue for the stakeholders, however, the cost needed to create a movie is very high, so it becomes a tremendous challenge for the production companies to produce movies that are highly successful and generates high revenue called blockbuster or a movie with the revenue bigger than $100.000.000. The competitive advantage the production companies currently have is the current advancement in technology, which if they are able to fully take advantage of it, they will be able to reduce the chance of producing failed movies. In this paper, artificial neural network is chosen as a method to predict the success of the box office movie. The ability of the artificial neural network method in making predictions certainly needs to be supported by determining the important variables that will be used in the construction of the model. In this paper, the author uses the chi squared test method for determining important variables of categorical input data and categorical output data and the ANOVA method for determining important variables of numerical input data and categorical output data. As a result, the authors managed to get an accuracy of 86% in the test set, which means that 86% of the test data can be classified correctly.