Movie Recommendation System using Content Based Filtering

Abstract

A recommender system is a tool that assists consumers in finding content and overcoming information overload. It predicts user interests & gives recommendation based on the user interest model. The existing content-based recommendation system is a collaborative filtering system under continuous improvement that does not require movie user testing. Instead, the similarities are calculated based on the knowledge of the movies the users selected and then making recommendations accordingly. With the development of machine learning, the current content-based recommendation system can create user- profiles and movies, respectively. They are creating or updating a profile based on analytics of user-friendly movies. The system can compare user and movie profiles and recommend the most similar movies. Therefore, this recommendation method that directly compares the user to the movie cannot be delivered to the co-filtering model. There are many features released in the movie; they are varied and unique and also different from other recommendation programs. Simply put, systems can suggest movies based on the person’s two or more attributes.