With the development of electronic commerce, recommendation system has become one of the most important technologies. To solve the black sheep users problem in collaborative filtering recommendation systems, social network analysis approaches is applied:Through calculating degree centrality of each node, the dataset is divided into two parts of black sheep and non-black sheep based on the degree centrality of the users. This study empirically shows that the characteristics of dataset can affect the performance of CF recommenda-tion systems;the method proposed can be useful for solving the black sheep problem.%随着电子商务的发展,推荐系统已成为最重要的技术之一。针对传统协同过滤推荐算法的“孤独用户”问题,提出了一种融合了社会网络分析的方法来解决此问题。即通过SNA的度中心性来将数据集分为“孤独用户”和“非孤独用户”组,不同的相似度测量和推荐方法被应用到两个数据组。实验表明,相对于传统的协同过滤算法,该方法能够显著提高推荐系统的性能,有效缓解“孤独用户”带来的问题。
展开▼