EMPLOYING AUTOENCODER IN BUILDING A MOVIE RECOMMENDATION SYSTEM

Abstract

We made a system that gives you suggestions based on what you like using a type of neural network. We use a model that has two parts to make suggestions based on what you like. Autoencoders are a kind of AI that can be used to make movie suggestions. This paper is about how to make computer systems that can recommend movies to you by themselves. We also talk about different kinds of autoencoders, like ones that remove noise, ones that keep only important parts, ones that learn to compress data, and ones that stack multiple autoencoders. The paper also looks at the good, bad and real-life stuff of using autoencoders for movie suggestions. This means that having lots of movie data with what people like and what the movies have is really helpful for autoencoders to learn from it. This paper tries to make movie recommendation systems better by using Autoencoders and testing different ways of doing things with them. Checking out the results from experiments on a public database shows how much better the usual ways of doing things are. In short, these systems can help you find movies that you like by using your own preferences and feedback.