Abstract: Collaborative filtering recommendation algorithms widely used in e-commerce, recommend interesting content for users from massive data resources by studying their preferences and interests. The focus of similarity and evaluation have been changed when applied to social networks, however, they cause low efficiency and accuracy of the recommendation algorithms. User similarity was introduced for redefining the attribute similarity and similarity composition as well as the method of similarity calculating, then a new collaborative filtering recommendation algorithm based on user attributes was designed and some methods for user satisfaction and quality of recommendations were presented. The experimental result shows that the new algorithm can effectively improve the accuracy, quality and user satisfaction of recommendation system in social networks.