Movie-Recommendation-Engine

Getting Started

This repository implements a Movie Recommendation Engine which recommends new movies to different users with FlunkSVD, Knowledge Based Recommendation and Content Based Recommendation.

Prerequisites

numpy
pandas

How to run the recommendation engine with the Recommender class?

Instantiate the recommender class

rec = r.Recommender()

To make recommendations for users in the dataset, call the make_recommendations function with first parameter as the user id and the second parameter as the keyword 'user'.

rec.make_recommendations(8, 'user')  

To make recommendations for users not in the dataset, call the make_recommendations function with first parameter as the user id and the second parameter as the keyword 'user'.

rec.make_recommendations(1, 'user')  

To make recommendations for a movie in the dataset, call the make_recommendations function with the parameter as movie id.

rec.make_recommendations(1853728)  

To make recommendations for a movie not in the dataset, call the make_recommendations function with the parameter as movie id.

rec.make_recommendations(1)  

Summary

Here, we have developed a Recommender System which uses FunkSVD to make predictions of user movie ratings. And uses either FunkSVD or a Knowledge Based Recommendation (highest ranked) to make recommendations for users. Finally, if given a movie, the recommender will provide movies that are most similar as a Content Based Recommender.