The overall objective is to build a Movie Recommendation system which is a Hybrid Recommender system. The steps will include implementing a few recommendation algorithms such as content based and collaborative filtering separately and then combining them to build our final hybrid recommender system. The final system recommends movies to a particular user based on the estimated ratings that it had internally calculated for that user and thus will have the quality of being more personalized and tailored towards particular users.
About the data: The dataset we have used in this project is ‘The Movies Dataset’. This dataset contains metadata for all 45000 movies listed in the ‘Full MovieLens Dataset’. The dataset has files containing 2.6 million ratings from 270,000 users for all 45,000 movies and all these files are in csv format. The ‘movies_metadata.csv’ file has information about the genre, popularity, rating, languages and many other features for each movie. The credits file consists of cast and crew information for all the movies. The ‘keywords.csv’ file has information regarding the movie plot keywords for the movies. The ‘rating_small.csv’ has information about the subset of 100,000 ratings from 700 users on 9,000 movies. The ‘links.csv’ file has information about the IMDB and TMDB IDs of a small subset of 9,000 movies of the full dataset. The last file called ‘links_small.csv’ consists of a smaller version of all the parameters that are already present in the links file, it contains 9099 movies.
Link to data: https://www.kaggle.com/rounakbanik/the-movies-dataset