implementation of a recommender system using collaborative filtering

Nowadays, consumers have a lot of choices. Electronic retailers offer a great variety of products. Because of this, there is a need for Recommender Systems. These systems aim to solve the problem of matching consumers with the most appealing products for them. They do this by analyzing either the products information details (Content Based methods) or users social behavior (Collaborative Filtering). This paper describes the Collaborative Filtering technique in more detail. It then presents one of the best methods for CF: the Matrix Factorization technique. Next, it presents two algorithms used for matrix factorization. Then, the paper describes the implementation details of a framework created by us, called Rho, that uses Collaborative Filtering. In the end, we present some results obtained after experimenting with this framework.