This project built for movie recommendation system based on collaborative filtering user based and item based recommendations.
Functions for recommendation written with MATLAB. I used MovieLens dataset to train model. Pearson correlation used for user based similarity and cosine similarity used for item based similarity. For improving the efficieny, I used significance weighting. I used 10fold to find best k values for each prediction method. To take recommendation from website, firstly you should register and login the system.
After that, you must rate 20 movie to get prediction.