Built a CF Recommender system to provide reasonable prediction of active user's preference on the basis of selecting & aggregating other user's opinion on MovieLens data set with a 16% improvement over base model leveraging & optimising User-User CF, Item-Item CF, suprise library, multiple Similarity functions, Averaging Functions & Neighbourhood size.