This is a recommender engine which recommends best movie for given user from the MovieLens data set.
- I have used 100K dataset provided by MovieLens for testing and training purposes.
- This dataset provides 100,000 ratings (1-5) from 943 users on 682 movies.
- Each user has rated at least 20 movies.
- I have used Graphlab-Create and Scikit Learn as Machine Learning libraries.
- Recommendation works using item-item similarity based method that ranks an item according to its similarity to other items observed for the user in question.
