Movie recommendation app with cosine similarity and flask

Abstract

People are constantly busy with their professions, businesses, and other endeavours in the world in which we currently live. The majority of them find that watching movies is the greatest way to unwind during the limited free time they have between jobs. However, with so many movies available in different languages, choosing which one to watch may be a time-consuming task. The recommendation may be collaborative filtering or content-based. As the name implies, collaborative filtering bases its filtering method on the interactions between relevant user behaviour and that of other users. In order to correlate movies with recommendations, this study describes cosine similarity. To receive the top 5 recommendations, the user must enter the name of a movie they have previously enjoyed. We've also introduced a feature that allows the user to optionally provide the year data that they want the engine to use to propose movies that were published after that specific year. For example, if the user wants to filter recommendations for new movies, they can provide a recent year.