User-adaptive Item-based collaborative filtering recommendation algorithm
Wang Cong-me · Jisuanji yingyong yanjiu · 2013
The traditional Item-based collaborative filtering algorithm regards every rating as equal importance when calculating the similarity between items,and ignores the impact of the similarity between co-rated users( users co-rate both two items) and target user on the similarity between items. This paper proposed a user-adaptive Item-based collaborative filtering recommendation algorithm,in which the rating of a co-rated user on an item was weighted by the user similarity between the co-rated user and target user,in order to select different neighbors of a certain target item for different target users,and so as to improve the recommendation accuracy. The experiment results suggest that the proposed algorithm can efficiently improve the recommendation quality.