Recommender systems helps users to choose the best option from the range of available items. The performance of these systems is based on the users' previous preferences or similarity of users. As a result, if information of the user preference is increased, the items proposed by these systems will be closer to the user's preferences. It should be noted that factors such as increasing the number of items or a lack of information about the user's preferences can make recommendations presented by RSs not close to the user's preferences. To avoid increasing the RSs error in the proposal we must obtain accurate information from users' tastes. In this article presented solutions solve the problem of recommender systems, such as sparse matrix.
