kaggle_movie_revenue_prediction

In this competition, one was presented with metadata on over 7,000 past films from The Movie Database predict their overall worldwide box office revenue. The data points had cast, crew, plot keywords, budget, posters, release dates, languages, production companies, and countries. Since the data is mostly tabular, with variable in a dictionary that can be extracted as a vector, I used a Random Forest Regressor to predict the revenue from most lucrative features. The competition was held a while ago and I referred multiple posted solutions to use the best possible feature engineering. This solution beat the rest of the solutions in predicting the revenue of the movies of provided test data.