Movie Recommendation using Content Based Filtering

Abstract— We propose a recommendation system based on machine learning that recommends movies to users based on movie metadata. The purpose of movie recommendation systems is to help movie viewers by suggesting films to watch without making them go through the difficult and time-consuming process of selecting from a large selection of films that number in the thousands or millions. The system takes an input movie and returns the top 5 recommendations based on the similarity of features such as genres, director, and cast. The system employs a simple similarity metric to compare the input movie with the recommended movies.