Movie Recommendation System Using Hybrid Approach

Abstract

In this digital era, there is enough content on the movie suggestion system already. The customer won’t have to spend a lot of time looking for stuff they could enjoy if movie suggestions are provided. To get user-specific movie suggestions, a recommendation system for movies is also essential. It has been discovered, after conducting extensive online research and consulting a large number of research papers, that the suggestions produced utilising Content-based filtering in order to compare the similarity of the vectors, filtering only employs a single method for text-to-vector conversion. To produce the final recommendation list for this study, the output of various text-to-vector conversion algorithms was altered and employed a variety of text-to-vector conversion strategies. It solely employs the content-based filtering technique, although it may be thought as a hybrid strategy. With the help of movie recommendation systems that employ content-based filtering, these systems can suggest a movie to the viewer depending on the content. People now use movies as a getaway from their hectic life. A big dataset of movies that are accessible globally makes it difficult to pick just one movie, making watching movies more difficult than it has to be. In this work, Heroku deployment has been used to create a recommendation system that used a content-based filtering method. The most common methods for creating recommendation systems are collaborative filtering and content-based filtering. The application may be easily deployed thanks to Heroku deployment.