Skip to Main Content
Collaborative filtering recommender systems which automatically predict preferred products of a customer using known preferences of other customers have become extremely popular in recent years. Recommending products based on similarity of interest is also attractive for many domains such as books, CDs, movies, etc., and reducing the information over load in the electronic commerce environments. The growth of customers and products in recent years poses some key challenges for nearest-neighbors collaborative filtering. Performing many recommendations per second for millions of customers and products becomes poor. Many algorithms proposed so far, where the principal concern is recommendation scalability, may be too expensive to operate in a large-scale system. This paper analyses the scalable collaborative filtering using clustering technology. This approach can implement with two ways. One is based on the user clustering technology and the other is based on the item clustering technology. There is also a hybrid method using the user clustering and item clustering or bi-clustering.