基于用户信任和兴趣的概率矩阵分解推荐方法

摘要

传统协同过滤推荐算法存在数据稀疏性、冷启动、新用户等问题.随着社交网络和电子商务的迅猛发展,利用用户间的信任关系和用户兴趣提供个性化推荐成为研究的热点.本文提出一种结合用户信任和兴趣的概率矩阵分解(STUIPMF)推荐方法.该方法首先从用户评分角度挖掘用户间的隐性信任关系和潜在兴趣标签,然后利用概率矩阵分解模型对用户评分信息、用户信任关系、用户兴趣标签信息进行矩阵分解,进一步挖掘用户潜在特征,缓解数据稀疏性.在Epinions数据集上进行实验验证,结果表明,该方法能够在一定程度上提高推荐精度,缓解冷启动和新用户问题,同时具有较好的可扩展性.%The traditional collaborative filtering recommendation algorithm has such problems as data sparseness,coldstart and new users.With the rapid development of social network and e-commerce,how to provide personalized recommendations based on the trust between users and user interest tag is becoming a hot research topic.In this study,we propose a probability matrix factorization model (STUIPMF) by integrating social trust and user interest.First,we excavate implicit trust relationship between users and potential interest label from the perspective of user rating.Then we use the probability matrix factorization model to conduct matrix decomposition of user ratings information,users trust relationship,user interest label information,and further excavate the user characteristics to ease data sparseness.Finally,we make experiments based on the Epinions dataset to verify the proposed method.The results show that the proposed method can to some extent improve the recommendation accuracy,ease cold-start and new user problems.Meanwhile,the proposed STUIPMF approach also has good scalability.