Garo Sarajian

This is one of my larger projects where I explore various recommender systems for movie recommendations. All of the analysis uses data from MovieLens dataset on Kaggle and is based on what I learned from The Lazy Programmer's online course on Udemy.

These approaches don't factor in individual tastes and are helpful for determining overall rankings and approaching the cold start problem.

An approach that dampens the mean rating based on the number of ratings.

An approach that ranks based on the lower bound of the confidence interval for the mean.

These approaches tailor recommendations to users. Stay tuned for posts on collaborative filtering and matrix factorization approaches.

RecFlix