Overview / Usage

Recommender System systems are the most commonly used concept for suggesting or giving user their their preference of their choice.Its been extensively used in e-commerce applications,browsing through web ,etc. Online media streaming portals like Netflix, YouTube ,AmazonPrimeVideos etc ,use the concepts of predictions and shows content of the stuff that users might like.Here we use the concept of Restricted Boltzmann Machines to predict the predict the like/Dislike of the user and Autoencoders to predict the rating of a new movie,which the user hasn't watched.

Methodology / Approach

The given dataset includes movie dataset of over 5000 movies which have been rated by over 1000+ users.First we Importing the dataset of movies,users and ratings ,now we Prepare the training and test set by converting them into an array.
Now we Obtain the number of users and movies we Convert the data into array with users in lines and movies in columns
Find the ratings into binary(0/1) Liked or Disliked,Now continuing from rearranging we use autoencoding to predict the range between the rating of the user might turnout.

Technologies Used

Language
● Python

Machine Learning Libraries
● TensorFlow
● Pytorch
● numpy
● pandas

Repository

https://github.com/TEJASNARAYANS/Restricted-Boltzmann-Machines