Recommendation Engine

This is a movie recommendation engine that has been implemented and deployed as a Flask application. It uses SQLAlchemy on the inside. There are two classes of algorithms -- one that uses matrix factorization like (SVD) and the other that uses clustering algorithms based on various different metrics (euclidean, cosine etc).

The dataset being used for the movies is the MovieLens Dataset

Installation and running-

cd RecomEngine
pip install -r requirements.txt
python app.py

Now open the devolopment server

Similarities

You can find similar movies to the ones in the dataset by navigating to the similarites parts and entering a MovieID from the dataset The clustering algorithm will find the closest neighbours and return them to you

Predictions

Just enter some ratings for movies by entering the Movie-ID and ratings and then press "Predict Ratings" This will then run a weighted average of the ratings of all similar movies for every movie and SORT the results, and present to you the hgihest rated movies