Movie Recommendation System Using Machine Learning

Abstract

In today's incredibly busy environment, recommendation systems are becoming more and more crucial. The most frequently used areas for recommender systems are, among other things, books, news, articles, music, videos, and movies. I have suggested a movie recommendation system in this paper. The importance of recommendation systems can be attributed to their ability to help people make the best decisions without having to use their cognitive faculties. Due to the numerous duties that need to be completed in the 24 hours available, people are constantly pressed for time. Like any other kind of recommendation system, a movie recommendation system relies on the similarities between users (collaborative filtering) or the user's desired activity (content-based filtering) to provide suggestions. By fusing collaborative and content-based filtering, we can get beyond the limitations of both approaches and build a more robust recommendation system. In this research, a movie recommendation system is presented that uses a hybrid model based on cosine similarity techniques and feature extraction. The list is ranked according to how highly members of the website rated each film. After that, it will be simple for the user to browse the suggestions and find their preferred movie. As a result, movie recommendations will be more pertinent to users' needs. The process culminates in a search of movie databases for all pertinent information, including popularity and beauty, needed for a recommendation.