The collaborative filtering algorithm based on domain ontology and user preferences

This paper proposes a method of user ratings similarity, based on forgetting function, which adjusts the importance of the user ratings according to time, considering the impact of the changes in user's preferences. While the user clustering algorithm is presented, which takes into account the personal characteristics of the user's information to affect its form ultimately choose to purchase goods weighting factor improved characteristics of the user similarity algorithm, reducing the nearest neighbor range of options.