This Big Data Analytics project aimed to utilise big data analytic approaches on R Studio to examine high volume of movie data to unravel hidden patterns, interrelationships, and gain other acumen. For multimedia platforms, collaborative filtering (CF) makes endeavours to help users retrieve their favourite films by detecting precise alike neighbours among users or films from their history patterns of shared ratings.
However, Because of data being sparse, the neighbour selection process for suggesting similar movies has become progressively complexed with the expansion of motion pictures and users growing exponentially. This project aimed to overcome this issue by proposing a real rating matrix and sparse matrix to narrow down diverse movie genres rated similarly by users on R studio. The experimentation on results on the IMDB dataset highlighted that User-based collaborative filtering outperformed Item-based CF. Considering the preceding statement, Item-based collaborative filtering (IBCF) model was executed to build a recommender engine with the purpose of recommending movies based on similar ratings amongst users, covering multiple genres.