A movie recommendation system based on the concept of collaborative filtering.
Have you ever imagined how OTT platforms recommend movies to us?
Sometimes, they ask us about our likings and use those along with some other predictors for the recommendation.
This project is a small form of how these recommendation systems work based on this paper of collaborative filtering.
The dataset used here is the Netflix prize money dataset.
The idea is to predict the ratings of the movies which are not yet rated by the user and based on the list of movies sorted in descending order of predicted ratings recommend the movies.
I have also used the concept of k-nearest neighbors algorithm ,where-in I am considering a set of users similar to any particular user and reporting the RMSE based on that.
Download the zip file or clone this repository and open the jupyter notebook on your PC or google collab and run the program.